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

AF 447 Thread no. 4

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

AF 447 Thread no. 4

Old 8th Jul 2011, 16:30
  #1021 (permalink)  
 
Join Date: May 2011
Location: BOQ
Age: 79
Posts: 545
Likes: 0
Received 1 Like on 1 Post
Re: Asking what would the the second thing to do after looking at AoA value when hearing STALL STALL.
Quote:
Originally Posted by OK465
Is this a trick question?

Nope . So you have your AoA value, lets say it's Seven.

If that's the first thing you look at it must be key to the subsequent process that you plan to apply. So what do you do next? What is so important about the AoA number that the aircraft is telling you in your subsequent decision process. If you need to question the stall warning then why do you believe the actual AoA, it's coming from the same system....

Me, the first thing I'd look at is a horizon (an AoA indicator of sorts of course, but which also tells me whether my wings are level. Two birds with one stone....)

…And from Post #994:

Very simply.

The nose down inputs were transient. They were not of sufficient size/duration to cause the THS to move.

Quote: It's a pity that the stall warning sounded again after they made the correct (ND) input.

@PA 18 151

I think one would be reasonable to accept that if I’ve got an ALT2 jet in my hands (A/P & A/T not available in ALT2), that I should, in fact, be flying “attitude” as a control parameter when I hear the “Stall, Stall”. So I know the aircraft attitude because I "may have" put it there.

With an analog AOA indicator, I can instantly correlate where I am in the aircraft operating envelope with respect to AOA…and observe the AOA trend. "Stall, Stall" is just a preliminary warning that I'm headed that way or possibly risk inducing an accelerated stall. It's not a matter of "believing" one or the other.

More importantly

As is continually being addressed here, the forward SS input may have to be held for some period of time. If all I can reference is a repetitive “Stall, Stall” audio, I have no idea to what extent my recovery inputs are either effectual or not.

“Stall, Stall” doesn’t tell me whether I’m at 35-degrees AOA “locked” or SEVEN degrees AOA trending lower. Neither does a stick shaker. If I can see a trend indicating I’m making the right input, I get positive feedback to hold that input. This might be particularly useful with a Stall Warning which suddenly reactivates.

The analog/digital AOA indicator available on the PFD of some 737 NG’s sits just to the upper right of the attitude display. It would take an unnatural effort not to see and correlate both Attitude & AOA with the stick shaker going. (Someone posted a picture of it eons ago. It is not an additional instrument but an OPC addition to an existing display. And it is certainly not overwhelming in day to day normal flying.)

Someone (LW 50 I think) previously asked if AOA displays were available in airliners and were they useful. Are they useful?

Same answer as before…depends on the driver.
OK465 is offline  
Old 8th Jul 2011, 16:35
  #1022 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,328
Received 104 Likes on 51 Posts
Originally Posted by jcjeant
This is no need of long ND input fot change the THS deflection.
Any input (up or down) is followed (a response) by a movement of the THS ..up or down (in the flight law they were)
Are you sure about that ?
The BEA Note tells us that it took the 30s continuous NU Input to get the THS from 3 to 13°.

Where did you find evidence that it followed suite to every Elevator input ?
henra is offline  
Old 8th Jul 2011, 16:59
  #1023 (permalink)  
 
Join Date: Jun 2009
Location: VA, USA
Age: 58
Posts: 578
Received 0 Likes on 0 Posts
Are you sure about that ?
The BEA Note tells us that it took the 30s continuous NU Input to get the THS from 3 to 13°.

Where did you find evidence that it followed suite to every Elevator input ?
+1

I, for one, do not believe that is how the THS works. Otherwise every small SS input would result in a re-trimed aircraft (even though it would be a small change).
GarageYears is offline  
Old 8th Jul 2011, 17:08
  #1024 (permalink)  
bearfoil
Guest
 
Posts: n/a
henra

howdy. The BEA report "a" left, nose up input. Not continuous. Doesn't the 30s mimic a rate similar to Manual? IOW, .65 degree/second? That gives a total of 20 degrees NU, but with some pauses on the way up, why not a total of Ten total excursion in 30 sec.? How would the THS "know" when to calibrate its rate? Maneuvering and cruise are separate in what way? (To the THS)? Is it not "On" or "Off"? If in Auto, (it was), then it is the FCS that did the inputs, and if not in response to PF, then what was the "Goal"?

If, at a/p drop, the THS was at a trimmed 3 degrees (It's "starting" point), then ten degrees is the total excursion from cruise trim, and we know that the a/c did NOT immediately start to climb, per BEA. This lack of Pitch response is trying to say something, post PF input ("I have the controls").

If unresponsive to PF's initial inputs, would he not increase the input? If ten seconds later, the THS would now have trimmed seven degrees NU? What took the a/c so "long" to start "climbing"? Imagine that, a 150 ton a/c at 350 being "unresponsive". Any vertical acceleration that resulted in 7k fpm climb would be a belly cruncher, and he would have countered the zoom with ND, without question. What made the A340 reluctant to drop her nose in Caraibes? Two conscious and trained pilots would not allow such a climb, and if they could not control it, why?

It is in this climb that some mechanical and/or FCS transient seems to appear. One must grant to each "system" a basic trust in its abilities? To include the Pilots, eh?

Last edited by bearfoil; 8th Jul 2011 at 17:35.
 
Old 8th Jul 2011, 17:17
  #1025 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
The two unsolved puzzles for me are:-
1) Why the climb in the first place
2) Assuming the Captain was able to diagnose a stall, why nothing apart from power really changed all the way down. It is pretty obvious that no further significant recovery action was taken from 10,000' down or the impact would have been quite different - not that I think there would have been enough height.
BOAC is offline  
Old 8th Jul 2011, 17:23
  #1026 (permalink)  
bearfoil
Guest
 
Posts: n/a
BOAC,

1. Why is for Philosophers, eh? How? Did not the PF initiate it?

2. Could they have believed they were Nose Down, Oversped, and Unrecoverable by anything other than NU inputs? Could the PNF's sudden input @ 10,000 feet ASL have been a ND? Did he peep an horizon off to his left? Not that ND would have accomplished anything but another STALLSTALL, right? They already tried that.
 
Old 8th Jul 2011, 17:25
  #1027 (permalink)  
 
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes on 0 Posts
The two unsolved puzzles for me are:-
1) Why the climb in the first place
Extreme or (wrong) sensed airspeed....????
hetfield is offline  
Old 8th Jul 2011, 17:46
  #1028 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,178
Received 377 Likes on 231 Posts
bear, you take us back to trying to unravel

What nose attitude did the flying crew see?
What nose attitude did the captain see when he re-entered the cockpit?

The primary flying instrument to reference when you are flying in instrument conditions is the attitude indicator.

There were three possible displays for the captian to view.
There is no reported evidence of failed attitude indication.
We do not see BEA report something amiss with attitude indications from the voice info released.
Considerable attention is given to attitude information from the FDR in their description of what happened, just as THS position is given considerable attention.

IF nose down input induced stall warning (the second series) due to an alpha function reawakening (and alpha/airspeed restoration, system confidence, etc) it seems odd to me for a crew to interpret that the alpha driving that warning is a high speed buffet sort of warning. From what I glean in various downloaded pdfs on A330 systems, high speed buffet, or approach to it, doesn't evoke a stall warning alert. If you get too close to VMO, or try to get past it, the system tries to correct you. It doesn't want to get beaten up by high speed buffet any more than the pilots do. (Or so the programming can be characterized).

Does that initial pitch have an innocuous origin?

"The AP will disengage if the high-speed protection is active."

Some have asked if the AP initial disengage led pilots to believe that they were in HS protection ... which if AS wasn't reliable, they'd have no way to cross check.

It seems to be the consensus that Alt 1 Law was in operation, and HS prot looks to be available in Alt 1, but not Alt 2. If the capability degraded to Alt 2, robot isn't making inputs for HS protection.
(From Flgt Trn Manual ... alt 2 ... in the case of failure of 3 ADRs, no high speed protection. )

You can expect up to a 1.75 g input (nose up) from the robot if HS protection is called for. After many posts, and four different threads, and going over this again and again, nobody has solid evidence that this feature is where nose up came from. The Abnormal Law possiblity remains open, but the reported evidence of a nose up command from SS keeps cropping up.

Even with ND inputs followed by a stall warning, there remains the matter of attitude flying.

If your nose is up and your are falling, aren't you usually
either
stalled,
in a deliberate powered descent (no applicable here, their mission was "maintain altitude and course" at this point in the flight)
or
undepowered for your desired flight profile?

The power levers were last reduced to ~55%, so the third consideration must be rejected ... which takes you back to nose high and falling, you are stalling.

What did they see on their attitude display?

Last edited by Lonewolf_50; 8th Jul 2011 at 18:43.
Lonewolf_50 is online now  
Old 8th Jul 2011, 17:54
  #1029 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,328
Received 104 Likes on 51 Posts
Originally Posted by Mr Optimistic
When speed did get below 60kt at some time (which it must have done to inhibit warner) why did that alone not force change to abnormal attitude law of and by itself given the 'or'?
One possible explanation would be a prior ADR disagree, invalidating the speeds. If the speeds are considered invalid this could be an explanation why the System did not switch to abnormal law based on speed threshold.
This ADR disagree could result from asymmetric flow conditions to the different Pitots and thus large discrepancies when flow into one pitot was stalled while still attached in the other one(s).
henra is offline  
Old 8th Jul 2011, 17:59
  #1030 (permalink)  
bearfoil
Guest
 
Posts: n/a
LW 50

Thanks for the response. I vaguely remember some buzz early on re: no Horizon, or attitude prompt, and a long discussion re: to the option of an AH, (offered by Bus, to be placed upper left on Captain's side of Panel. Such option not selected on this 330 by AirFrance.

It is a seductive argument, this "What did he see"? It isn't particularly relevant, except to make a foundation for some future finding, to wit:

1. He saw nothing. Unlikely, because he did make manual input (BEA)

2. He saw something. Likely, see above.

3. He saw too much. This falls just short of being demonstrable, by discussion.

4. He saw just what he needed and made the most bone-headed call of his short 37 years on the Planet.

I wish the in fighting would stop, but I think it is merely reflective of how BEA and the other players are playing the Public.

The fact that I can still ask these simple questions, and not be accused further of "Bunkum", must mean that we have not progressed much further than where the discussion was after the ACARS leak.

Nowhere on the Planet can I avail myself of data such as is here. Helo pilot, (), 330 instructor, , Aero E, , etc. ad vol. BEA has made possible a political and economic discussion also, by being coy.

This is my idea of Heaven, but for the grim results of what will most likely eventuate as a mundane "told you so," by myriad posters.

henra. You mention "Asymmetric airflow to the Pitots". This is not new, and likely precisely what occurred. What do you suppose is the obsession with ICE, here? As in, turbulence caused a/p loss? Consistent with W/S and T/CAS alerts (for arguable reasons). Given P/F made instant input at a/p loss, and the a/c did not climb straightaway, turbulence seems to have some back-up evidence, whereas ICE?
 
Old 8th Jul 2011, 18:01
  #1031 (permalink)  
 
Join Date: Jun 2009
Location: Bedford, UK
Age: 70
Posts: 1,319
Received 24 Likes on 13 Posts
Or they could see their (true) attitude and interpreted the situation in such a way as to render nose up the best option. How about they had lost all faith in their plane and were of the opinion that it was not responding appropriately to their inputs, eg a major structural/system fault. When they tried nose down, the infernal stall warner was a shock and the last input was reversed to allow more thinking.
Mr Optimistic is online now  
Old 8th Jul 2011, 18:07
  #1032 (permalink)  
 
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes on 0 Posts
@AZR:

IIRC F/CTL SD page is automatically selected when certain types of failures occur. Was the case on 447, based on the ACARS. I think interim reports from the BEA mention that.
Correct, F/CTL page was called @ ~2:10:16 due to F/CTL RUD TRV LIM FAULT.
SD returns to normal Cruise page (if no other pages in queue) after confirmation by crew and clearing ECAM message.

Other faults present at this time don't call any SD page.

@ 2:11:40 'Capt enters cockpit', the Cruise page would be in view at the time. F/CTL page only if crew didn't cleared the messages. (not acc. protocol)

The other warnings which called F/CTL page are F/CTL PRIM 1 FAULT and F/CTL SEC 1 FAULT @ ~2:14:00 much later in the sequence.

Last edited by A33Zab; 8th Jul 2011 at 19:04.
A33Zab is offline  
Old 8th Jul 2011, 18:10
  #1033 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Hi BOAC,
Originally Posted by BOAC
The two unsolved puzzles for me are:-
1) Why the climb in the first place
2) Assuming the Captain was able to diagnose a stall, why nothing apart from power really changed all the way down. It is pretty obvious that no further significant recovery action was taken from 10,000' down or the impact would have been quite different - not that I think there would have been enough height.
1. Not only the climb but the persistence of NU imputs by the PF.
2. Everything recorded (leaks) by the CVR is pointing at the conclusion that it seems that nobody ever acknowledged the stall situation, Captain included. This time, I could give some credit to the leaked informations as they are comming from BEA officials during interviews.
Today, another press article: Le Figaro - France : AF 447*: le pilote n'a pas compris la chute
takata is offline  
Old 8th Jul 2011, 18:20
  #1034 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,328
Received 104 Likes on 51 Posts
Originally Posted by Lonewolf_50
If your nose is up and your are falling, aren't you usually stalled, or undepowered for your desired flight profile? Well, the power levers were last reduced to ~55%, so the latter consideration seems to have been rejected, so you are back to nose high and falling, rather than nose high and climbing.

What did they see on their attitude display?
Either that or they didn't trust the AI.
Maybe it contradicted their 'picture' of the situation and they considered it unreliable as well ? When bombarded with Warnings and Failure messages it may become difficult to sort out what exactly is still properly working and what not.

As a response to the question by a member here in the thread of what the pilots felt acceleration wise it has to be noted that they would have felt a normal vertical g and accelerating (slightly increasing) feeling of an order of magnitude of a take off acceleration (~ 0,2g) (Due to Pitch attitude).
Without any visual reference this could heavily contradict the preconceived notion of what you would feel in a stall.
(When training stalls in Light Aircraft at the beginning of the career you will have a ND attitude after the stall leading to a more decellerating/falling sensation rather than accelerating and normal vertical g). This together with no outside reference and dubious instrument readings might give you a misleading picture of what is going on.
Don't underestimate the power of the human 'Seat of the Pants' feeling on our perception.
henra is offline  
Old 8th Jul 2011, 18:26
  #1035 (permalink)  
 
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes on 0 Posts
@henra:

+1

NAV ADR DISAGREE needs 10s monitoring to trigger.

ACARS msg was received @02:12:51; CMC has 60s maximum to correlate
to any failure message ~02:11:51 - 10s monitoring = ~02:11:41.


"At around 2 h 11 min 40
, the Captain re-entered the cockpit. During
the following seconds, all of the recorded speeds became invalid and the
stall warning stopped."


A33Zab is offline  
Old 8th Jul 2011, 18:27
  #1036 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,328
Received 104 Likes on 51 Posts
Originally Posted by Mr Optimistic
Or they could see their (true) attitude and interpreted the situation in such a way as to render nose up the best option. How about they had lost all faith in their plane and were of the opinion that it was not responding appropriately to their inputs,
We can't exclude that.
But it would only make sense if there was a prior significant and persistent attempt to try the classic stall recovery.
In a 200t Airliner you don't give up a serious recovery attempt after 5s. It will for sure take a siginifcant amount of time to get such a thing flying properly again after really enetring a stall.
At least i fail to see any significant indication of that in the information bits we have at hand.
henra is offline  
Old 8th Jul 2011, 18:50
  #1037 (permalink)  
YRP
 
Join Date: May 2005
Location: Ontario, Canada
Posts: 163
Likes: 0
Received 0 Likes on 0 Posts
So now that the discussion is back to the THS position, I'd like to repeat my question from post 514.

It seems that the crew made little or no ND inputs (apart from briefly shortly after the captain's return). So the THS position was not have been a factor. Whether there is enough elevator authority to recover is only a factor if they actually attempted it, no?

(I'm not making a statement on whether or not there was enough authority.)
YRP is offline  
Old 8th Jul 2011, 18:52
  #1038 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,178
Received 377 Likes on 231 Posts
henra, thank you. (Note: I don't underestimate seat of the pants, I've had the leans (vertigo) more than once while flying. Hate it. It's got to be fought through, and the usual starting point is by getting the attitude indicator back into the scan and keeping it there.)

By classic stall recovery, are you referring to lowering pitch (slightly, or a lot) to reduce AoA, or the TOGA and set X attitude (5 or 10, based on high or low alt environment) recovery technique? Seems the latter was attempted. (TOGA selected at any rate ...)
Lonewolf_50 is online now  
Old 8th Jul 2011, 18:57
  #1039 (permalink)  
YRP
 
Join Date: May 2005
Location: Ontario, Canada
Posts: 163
Likes: 0
Received 0 Likes on 0 Posts
BOAC:

rrat "No one would design a system like you seem to suggest with" - your confidence is most impressive. Does it extend to knowing that no-one would design a system to shut off the stall warning in flight whilst the a/c is still stalled?
I think the problem here (system design wise) is what to do when the sensors are outside their reliable range.

Presumably the AoA sensors have some envelope in which they work. The system designers would not have turned them off for no reason. The most likely reason is that (for some reason) the AoA is not realiable at that airspeed. Or perhaps there is another failure mode they are worried about (don't throw stall warnings at the pilot if it could be because of a bad AoA sensor).

In this case, continuing the warning was the right thing to do (ie would have led to the least confusion). In other cases it might be wrong.

IMHO a better solution would be to put "?STALL?" on the display somewhere (ie tell the pilot "might be stalled but not sure"). It is hard to do when the normal stall warning is the verbal stall-stall though.
YRP is offline  
Old 8th Jul 2011, 19:09
  #1040 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,328
Received 104 Likes on 51 Posts
Originally Posted by Lonewolf_50
By classic stall recovery, are you referring to lowering pitch (slightly, or a lot) to reduce AoA, or the TOGA and set X attitude (5 or 10, based on high or low alt environment) recovery technique? Seems the latter was attempted. (TOGA selected at any rate ...)
Good Point !
I referred to the lowering of the pitch.
Indeed setting TOGA could be indication of a stall recovery attempt.
However, if that fails (TOGA) the next logical iteration I assume would be ND maybe in conjunction with thrust reduction before trying rather weird techniques like full NU.
henra 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.