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

AF447 Thread No. 3

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

AF447 Thread No. 3

Old 11th Jun 2011, 15:14
  #1801 (permalink)  
Thread Starter
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
CVR

It's not tennis.....
Roland-Garros or not, ChristiaanJ, do you know any French people saying so little in 4 minutes ... ?
CONF iture is offline  
Old 11th Jun 2011, 16:11
  #1802 (permalink)  
 
Join Date: Jun 2003
Location: UK
Posts: 3,053
Likes: 0
Received 29 Likes on 14 Posts
bearfoil,

Lex parsimoniae.

The picture is incomplete, but the gist is clear. The a/c did not behave badly, nor unexpectedly. Certainly not according to the report.

1. The PF for some reason nosed up and stalled.

2. In the stall, the corrective response was TOGA + nose up.

The answers as to why 1 and 2 happened are currently simply guesses. The clues will be in the CVR and the traces. I suspect that the answers will be much more "human" than technical - for example if the UAS was interpreted as a windsheer speed loss, then the low altitude response is TOGA + pitch up. That drill is always done at low level in the sim, and it is a regular. You know you are doing it, because it is as bumpy as hell and your speed just vanishes...

If you think that the IAS is erroneous, but you are convinced that you are flying quickly, then the lack of response to TOGA + NU will just be confusing. It's easy for us to say "well you weren't flying quickly, laddy", but if some other clue biased or confirmed his decision, then there was no escape.

The CVR will, as ever, be interesting.

I have said it before, but I'll say it again. More high altitude training is needed, including dealing with serious faults in nasty conditions. It simply isn't done at the moment.
HundredPercentPlease is offline  
Old 11th Jun 2011, 17:04
  #1803 (permalink)  
bearfoil
Guest
 
Posts: n/a
HundredPercentPlease

With you.....100%. I framed it differently, but I did include that the a/c behaved as it was programmed to do. One can squeak a greaser in the Navy and get no more than "satisfactory".

One should not expect applause for doing what one is expected to do. The problem is almost certainly where you indicate it to be.

Are their mitigations? youdambetcha........

Responsibility is the issue, not Blame. Should the pilots develop their own paid for syllabus (pun intended)? To include coverage of events that happen once in a blue Moon? Some one has to. We have here Pilots who could write a training manual, and improve the performance of aviation. Simply reading their critique of the 'system' shows that. Tradition would have it that this Manual would not be 'selected'. The final moments of 447 were a Trade, Experience for Programming, and the result was a breached contract. As with all contracts, this one will be also 'unwound', the salient soft spots hardened with some resin, and push, please.

This time, I am hoping that the flying public, and aviators, will be better served by the ones traditionally expected to 'Fix It'.
 
Old 11th Jun 2011, 18:28
  #1804 (permalink)  
PJ2
 
Join Date: Mar 2003
Location: BC
Age: 76
Posts: 2,484
Received 0 Likes on 0 Posts
DJ77; Since I cannot find an answer in the material available to the question you and others have asked, "What Garage Years has been asking for a while now is the value of the critical AoA (AoAc) which is used by the confusers to trigger the stall alarm in ALT law when airspeed (including Mach number) is unreliable.", I am wondering if the actual stall AoA is a function of the individual aircraft with the actual numbers supplied with the fin. We know that in Normal Law, the stall warning is inhibited at < 23deg AoA. In any case, here is something. Note that it is not the stall warning which is unavailable below 60kts but the airspeed, which is considered to be invalid data.

AOA computationThe A/C is wired in the AOA unique selection, so the ADR computes the AOA with the reading of one resolver. The second resolver is used as a back-up.

The ADR determines the corrected angle of attack value which depends on the system number and the SLAT/FLAP position.

For a CAS less than 60 kts: AOAc = 0 and status matrix is coded NCD [no computed data].

For a CAS more than or equal to 60 kts, the corrected angle of attack is calculated as follows:
AOAc = (AoA ind / K) + where:

AOA ind = Indicated Angle of Attack; K, I = Correction coefficients which are a function of system number, slat and flap configuration and magnitude AOA indication.
PJ2 is offline  
Old 11th Jun 2011, 19:10
  #1805 (permalink)  
bearfoil
Guest
 
Posts: n/a
Hi PJ2, thanks for all you have added here. I have a simple, perhaps an even simplistic question(s), hoping you have the time to reply.

Re: your last post, what are your thoughts relative to LAW and the STALLSTALL @ a/p drop, and after PF's first control input? Had the a/c entered ALAW 1 or 2 at that point? If so, wouldn't the Alert have been somewhat close, (following) the a/p loss alerts? Would the AoA have been responsible for the Warning? If so, could the AoA have been influenced by either UP or DOWN Flow? Could the AoA have been (transient) in either very Low or very High values?

Is there any chance that the a/c could have been in Normal Law long enough for the Alert to have been triggered by greater than AoA 23 degrees? Or, if a DOWN Flow, a negative AoA?

If in ALAW 1 or 2, Would the AoA have been indicating legit? If so, how could it then silence (Cricket, STALL) and allow for a somewhat large increase in PITCH? Also a climb to 380? At the zenith of the climb, when PITCH 16, and AoA 6, instead of decreasing, could the AoA have actually been increasing to +6? I know many would appreciate your thoughts...

Last edited by bearfoil; 11th Jun 2011 at 19:20.
 
Old 11th Jun 2011, 19:26
  #1806 (permalink)  
 
Join Date: Jun 2009
Location: florida
Age: 81
Posts: 1,610
Received 55 Likes on 16 Posts
Caution, warning

From this old curmudgeon, I would wait a bit more until we can look at second-by-second replays of all the data.

Surely, several of us here have misgivings about control law reversion and cockpit displays and faulty sensors. No big deal.

I would not try to recreate the actions of the crew, especially after we have now seen many of the criteria for displaying caution and warning indications to the crew.

just some thots from an old fart......
gums is offline  
Old 11th Jun 2011, 20:26
  #1807 (permalink)  
 
Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,777
Likes: 0
Received 0 Likes on 0 Posts
Cool

Hi,

The solution ??
HeadUpFlight.net
jcjeant is offline  
Old 11th Jun 2011, 20:33
  #1808 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Hi Bear;

At its zenith ...
The airplane’s altitude reached its maximum of about 38,000 ft, its pitch attitude and
angle of attack being 16 degrees
.
At that point the v/s became rapidly -10k ft/min and the AoA continued to increase except for a short period when ND inputs were made and the SW sounded again.
mm43 is offline  
Old 11th Jun 2011, 20:36
  #1809 (permalink)  
 
Join Date: Aug 2008
Location: Betwixt and between
Posts: 666
Likes: 0
Received 0 Likes on 0 Posts
Hyperveloce,
It was a note (not a report, to be published before the end of july) released under the pressure of leaks, speculations, and of Airbus (which wanted the preliminary data to be published before the Bourget air show).
Symantics. It is reporting events, it is a report. The fact you say it was released under pressure from Airbus before an airshows says that this is intended to be a damage limitation exercise, which is pretty damning and would confirm my worst fears about the politics. It is then clearly apparent that the intention is to get the pilots to take the blame, as evidenced by the lack of clarity and spin in the report.
Sciolistes is offline  
Old 11th Jun 2011, 20:57
  #1810 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
jcjeant,

Thanks for the HeadUp Flight link. I read the article relative to who is in command, the pilot or the computers. It is a worthwhile read, like anything, some will agree, some will not, but it is food for thought.

The article can be found here:
Pilot or Computer in Command
Turbine D is offline  
Old 11th Jun 2011, 21:03
  #1811 (permalink)  
bearfoil
Guest
 
Posts: n/a
mm43

Thanks much. So one can assume that acceleration down from the top was pretty much non-aero? For a visual, would it be correct do you think to say that the tail was falling faster than the nose? (At first, AoA from 16 to 35 degrees, til impact?)

Also, that the increasing AoA during climb may have had little to do with controls, more to do with loss of velocity?
 
Old 11th Jun 2011, 21:15
  #1812 (permalink)  
 
Join Date: Mar 2008
Location: Malvern, UK
Posts: 425
Received 1 Like on 1 Post
Sciolistes

It is reporting events, it is a report.
Actually I have rarely read an aircraft accident report more loaded with innuendo.

A great deal is made of

the inputs made by the PF were mainly nose-up
however despite...

the PF made pitch-down inputs. In the following moments, the angle of attack decreased, the speeds became valid again and the stall warning sounded again
...it is perhaps more significant that

The angle of attack, when it was valid, always remained above 35 degrees
I believe it is likely they had insufficient pitch authority to recover from this. The only question is whether there where any other things they may have tried. However the lack of detail between 2 h 12 min 02 and 2 h 14 min 28 means we do not know.
Dont Hang Up is offline  
Old 11th Jun 2011, 21:19
  #1813 (permalink)  
 
Join Date: May 2011
Location: venice, ca
Posts: 61
Likes: 0
Received 0 Likes on 0 Posts
Knowing it all

JD-EE

Just to clarify: Regarding a pilot being "Released" for flight, no where would I indicate anyone "knows it all" or is a seasoned pro. But instead is merely qualified. Period. Hopefully as you said one will continue to learn and grow.
But you can't also make the case based on merely numbers of hours.

With the extremely limited bits of information on conversations between the pilots no one really knows what went on and what happened amid all of the confusion.
wallybird7 is offline  
Old 11th Jun 2011, 21:49
  #1814 (permalink)  
 
Join Date: Jun 2009
Location: France
Age: 83
Posts: 41
Likes: 0
Received 0 Likes on 0 Posts
The fact you say it was released under pressure from Airbus before an airshows says that this is intended to be a damage limitation exercise
Yes. Limitation, or prevention of damages, by media speculations, including the internet.
milsabords is offline  
Old 11th Jun 2011, 21:49
  #1815 (permalink)  
 
Join Date: May 2011
Location: venice, ca
Posts: 61
Likes: 0
Received 0 Likes on 0 Posts
Feeling the speed

Regarding feeling the speed, every pilot knows that if you're eyes are closed and you jiggle the stick, you cannot tell anything regarding speed, attitude, or up or down. It is called VERTIGO.
Trust your instruments, and if they don't work then you got trouble.
wallybird7 is offline  
Old 11th Jun 2011, 21:59
  #1816 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Bearfoil;
For a visual, would it be correct do you think to say that the tail was falling faster than the nose?
The following graphic (an adaption of one originally posted by Tailspin Turtle) should help with the conception. Note that all the speeds are in terms of GS and that the THS angle of incidence is 13° less than the nominal AoA experienced by the wing.



The 13° NU position of the THS will most likely have helped in stabilizing the descent - see post #1393 by PickyPerkins and post #1395 by grity.
mm43 is offline  
Old 11th Jun 2011, 22:03
  #1817 (permalink)  
 
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes on 0 Posts
Stallstall

The complete STALL WARNING Tables:
Typed it over and copied as image, so no warranty on typo errors!!

A33Zab is offline  
Old 12th Jun 2011, 01:02
  #1818 (permalink)  
PJ2
 
Join Date: Mar 2003
Location: BC
Age: 76
Posts: 2,484
Received 0 Likes on 0 Posts
bear;

I'm a data guy and gums is right - a detailed analysis isn't wise nor is it possible. I think dissecting the BEA Note word-by-word and looking for subtle meaning between the lines is a mug's game. We'll have a lot more in less than a month and they know that. They'd choose their words carefully of course but eventually it will all be there to examine.

Re your questions:

Re: your last post, what are your thoughts relative to LAW and the STALLSTALL @ a/p drop, and after PF's first control input?
The a/p dropped out due to NCD [No Computed Data - airspeeds]. As a result of the NCD, the flight laws reverted to Alt2 as designed. No control is lost, no squirrelly behaviours lurking, just an airplane with the same flight controls as before. No big deal. Really. Ailerons respond the same way, pitch is gee-driven with "limited feedback and gains." If one is accustomed to the airplane, hand-flying it at FL350 is a non-issue providing one is gentle and makes tiny corrections. The airplane WILL respond aggressively (as it is designed to) if large, swift inputs are commanded. Take it from there.

Had the a/c entered ALAW 1 or 2 at that point? If so, wouldn't the Alert have been somewhat close, (following) the a/p loss alerts?
The Monitoring process of the paired, "Command/Monitor" design of all flight computers provide input to many other computers including the FWC which distributes the messages to the CMC (ACARS) in the manner described in the first BEA Interim Report.

The timings of events are details and may not even be specifically recorded at the moment they occurred, though their occurrence is recorded. As with the ACARS messages when we saw them for the first time in July 2009 or so I don't think we can take very much from the order of failures.

Things were unfolding rapidly and it takes time for other things to catch up.

In part, that's why its important to wait for things to settle down, and, other than maintaining controlled flight by maintaining pitch and power before the failure, see how the loss of airspeed unfolds.

Very little by way of aircraft behaviour or external environmental factors when flying a transport aircraft requires immediate, instant response from the crew. The pilots here could name the few that do require swift, accurate action, and the UAS is not one of them at cruise altitude. A TCAS avoidance event should never be noticed in the back, for example.

Would the AoA have been responsible for the Warning?
The stall warning is AoA-driven as is now known, but as per A33Zab's excellent and very helpful post above, the AoA at which the stall warning triggers is the summation of Tables A & B and where applicable, C. (I don't know what the 2 and 3 digit numbers mean in the Slat-Flap Table in the table below Table A.) Other than what the tables indicate, the airspeed (CAS & Mach) do not have any link to the stall warning system.

If so, could the AoA have been influenced by either UP or DOWN Flow? Could the AoA have been (transient) in either very Low or very High values?
Very momentarily, perhaps, but the variations would be tiny - half a degree, etc. I've accessed the "Alpha Lab." parameters page in the ACMS menu many times and watched the AoA as well as the VRTG and LATG while in turbulence. The AoA varies around a pretty constant reading of 2.3deg NU or so, (>3 would be unusual, in my experience), depending upon weight. There is no evidence of a significant, sustained up/down draft in what has been provided by the BEA. While the aircraft was in unstalled flight, I doubt very much whether low or high values would obtain at all by up/down draft - weather. One minute before the first event, no one was especially excited or concerned about the plan to divert around weather. Not impossible, but unlikely/improbable.

Is there any chance that the a/c could have been in Normal Law long enough for the Alert to have been triggered by greater than AoA 23 degrees? Or, if a DOWN Flow, a negative AoA?
No on both questions. The pitch up did not begin until after the reversion.

If in ALAW 1 or 2, Would the AoA have been indicating legit?
We have to grant that the AoA data from all three sensors was working normally in the three hours before the event. Finding/seeing an abnormality in the data would not be difficult, even given the small changes; such would likely have been mentioned in the BEA Note. On the reversion to Alternate Laws, the flight control laws have nothing to do with the AoA-Stall warning system.

If so, how could it then silence (Cricket, STALL) and allow for a somewhat large increase in PITCH? Also a climb to 380?
Please reference the very fine work done by mm43 - an excellent presentation I think, then in the mind's eye, rotate the airplane about its center, imagine high-speed airflow from left-to-right and imagine what that flow does at the surface of the wing, and how the angles he has drawn change. Not to simplify but to emphasize, airflow around the "interruption" of an aircraft wing into airflow, is a "living" process, always changing, always in motion, always balancing the forces in response to where it is pointed and where it is actually tracking. (That is the value in the FPV on the PFD...used it many times to do steep turns in the sim! - they may have tried it to verify what would have been a "beyond-belief" rate of descent).


Specifically, Pitch and AoA are not linked so a conclusion that the pitch was "higher" does not necessarily link to a higher AoA which had originally triggered the stall warning.

Pull a bit too quickly on the stick even for a second or two and the AoA can go from 2.5deg to 6 very quickly while the pitch remains essentially level as the huge mass (airplane) catches up in another second or two.

High altitude flight is very different than work down lower. As Davies points out and as those who do this work know, the damping effects upon the aircrat of thinner, (at 350, less than a quarter of the density at SL), air are far less effective. He specifically, clearly, cautions that one must be gentle with the aircraft. That hasn't changed in fourty years whether cable-and-pulley or fbw with protections.

At the zenith of the climb, when PITCH 16, and AoA 6, instead of decreasing, could the AoA have actually been increasing to +6?
That's a question which must be answered by the data. We cannot go by how the text is written in the BEA Note - the data's the thing. We know the AoA rapidly ended up at 16deg just after the zenith and increased from there.

There is nothing "new" in any of this. It is standard stuff learned in courses on high altitude flight, weather, aircraft systems and so on as part of the craft. As always there will be nods and disagreements but the general principles hold.
PJ2 is offline  
Old 12th Jun 2011, 01:22
  #1819 (permalink)  
bearfoil
Guest
 
Posts: n/a
PJ2

A bonanza, and I am most grateful; I am sure others are also. I note especially the need for gentle handling at altitude!
 
Old 12th Jun 2011, 01:56
  #1820 (permalink)  
 
Join Date: Aug 2008
Location: Betwixt and between
Posts: 666
Likes: 0
Received 0 Likes on 0 Posts
Milsabords,
Yes. Limitation, or prevention of damages, by media speculations, including the internet.
If an accident investigation is manipulating the release of information in order to manage the debate on a particular accident, then by definition there is is an intention to subvert the message. This is quite clearly apparent in this BEA report and the subsequent news coverage. Therefore, in effect the effect of the report is to create speculation, in this case speculation over the implied sub-standard performance of the pilots.

If there were no interim or update reports of this nature, then there would not be any, meaningful speculation in the media as the public would be aware the final and most complete assessment practically possible will be released in due course.

It is extremely concerning that people justify the release of that BEA update, clearly worded to divert attention away from a sub-standard aircraft configuration ahead of a major sales event! I just don't understand how this can be acceptable.
Sciolistes 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.