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

AF 447 Thread No. 11

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

AF 447 Thread No. 11

Old 21st Nov 2013, 19:43
  #881 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Peter H;

You mean these two events?

1.16.3 Analysis of functioning of systems
1.16.3.1 Analysis of the initial sequence
Analysis of the FDR parameters and of the data contained in the two FMGECs’ nonvolatile
memories showed that:
ˆˆ The ADR 2 speed fell between 2 h 10 min 03.5 et 2 h 10 min 05;
ˆˆ the ADR 1speed fell for less than one second from 2 h 10 min 04 s to 2 h 10 min 05,
causing:
yy the disconnection of the autopilot,
yy the triggering of “PITOT PROBE” monitoring in the FCPC causing the transition
to alternate 2B law;
HazelNuts39 is offline  
Old 21st Nov 2013, 21:00
  #882 (permalink)  
 
Join Date: Oct 2013
Location: Wengen
Age: 53
Posts: 380
Likes: 0
Received 0 Likes on 0 Posts
PIO

I agree that the roll oscillation was a PIO.
I'm not so convinced that there is a big difference between alt 1a and 1b, though there certainly is some.
In alt 1a the airplane would have been inherently more roll stable and probably would not have required much correction to begin with but given a 3/4 stick input in reaction to an 8° bank angle (exccessive), a PIO is still possible
Winnerhofer is offline  
Old 21st Nov 2013, 21:09
  #883 (permalink)  
 
Join Date: Jun 2006
Location: B.F.E.
Posts: 228
Likes: 0
Received 0 Likes on 0 Posts
Personal opinion in response to the earlier post: in considering the possibility that the PF was so preoccupied with a roll PIO that he basically "tunnel-visioned" out his pitch situation (due to ALTN 2 law providing almost double the roll rates of ALTN 1 or normal law), his last comments on the CVR contradict that scenario. He states that he had the stick back the entire time, implying that that is exactly what he intended to do. Why he chose to do that we will never know. Whether his training or experience did not provide him the tools he needed, whether the design itself bears some blame, or whether he simply did not meet the challenge on that day.

One way or another, he used the best judgment he could manage at the time, and it steered him wrong. "But for the grace of God", any of us could be there in his place.
hikoushi is offline  
Old 22nd Nov 2013, 00:40
  #884 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
@A33Zab
!But that is not the issue!
Thanks for elaborating on the dependencies.
Owain Glyndwr in Thread No.8 - Page 47 -
.... one doesn't calculate AoA from CL but CL from AoA. CL doesn't figure directly in assessment of whether the aircraft is near stall or not - that is a simple function of AoA and Mach Number. If AoA is measured (as it is) then at low Mach that is enough to determine stall margin. When at higher Mach the boundary needs to be adjusted (which is why a simple AoA gauge cannot function as a stall warning when airspeed information is lost), but even then it would be possible to work with a default value which, even if it did not prevent "stall", would serve to avoid the sort of shambles that we see in AF447.
... described it similarly.

I've said it before, and I'll say it again - this PF never registered the Law change and its implications, neither in pitch or roll attitudes. Startled by the A/P disconnect and oblivious to anything else but dealing with the roll, he pulled back in the belief that AoA Prot would keep the aircraft in a stable longitudinal configuration. Revealing a deadly lack of knowledge in the FCS dependencies. Nothing changed over the next 4 minutes.
mm43 is offline  
Old 22nd Nov 2013, 07:35
  #885 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
It seems Dozy and Lonewolf are intent on dragging the 447 AoA issue into the Asiana thread so I will bounce it back here
Originally Posted by DW#167
No, no, no - the <60kts AoA NCD limitation was nothing to do with an air/ground assumption. It was simply an encoding of the vane manufacturer's specifications - i.e. that the vane could not provide reliable data without 60kts of forward airspeed.
My highlighting to show the flawed thinking which I assume existed at AB too.
BOAC is offline  
Old 22nd Nov 2013, 07:56
  #886 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Airspeed is derived from the differential between pitot and static pressures. Is there any airplane that derives it from static pressure alone?

Hindsight wisdom ...
HazelNuts39 is offline  
Old 22nd Nov 2013, 08:09
  #887 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi HazelNuts39,
Is there any airplane that derives it from static pressure alone?
All aircraft have a device which measures rate of change of static pressure - it's called the Vertical Speed Indicator.
AF 447 had in excess of 10,000 ft per minute displayed (about 100 kts) - more than enough for the "60kt AoA vanes" to sense I would suggest.
rudderrudderrat is offline  
Old 22nd Nov 2013, 08:26
  #888 (permalink)  
 
Join Date: Jun 2011
Location: West of Offa's dyke
Age: 88
Posts: 476
Likes: 0
Received 0 Likes on 0 Posts
BOAC

My highlighting to show the flawed thinking which I assume existed at AB too.
With all due respect to Dozy, using his terminology to infer AB's thought processes is rather a big assumption
Owain Glyndwr is offline  
Old 22nd Nov 2013, 09:22
  #889 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
rudderrudderrat,

please name an airplane that uses vertical speed as a substitute for airspeed
HazelNuts39 is offline  
Old 22nd Nov 2013, 14:18
  #890 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
rudderrudderrat- HN39 seems to have gone 'adrift' somewhere and I am totally confused by his reference to static pressures! Do we know which post he was referring to?

Owain - 'assume existed ' - why else pick '60kts of forward speed'?

On both counts we seem to be risking a trip to Oozlum land!
BOAC is offline  
Old 22nd Nov 2013, 15:13
  #891 (permalink)  
 
Join Date: Jun 2011
Location: West of Offa's dyke
Age: 88
Posts: 476
Likes: 0
Received 0 Likes on 0 Posts
BOAC

My point, not very clearly expressed, was really that 'forward speed' is an expression of Dozy's; I am sure that AB's engineers knew and know very well that it is the total air speed over the vanes that drives them not forward speed alone. [and so does Dozy I suspect, it was probably just loose wording]

In normal flight the two speeds are identical to all intents and purposes, although not of course when descending at around 10000 fpm (100 kts) and a forward speed which is also around 100 kts. Which is where the confusion between HN39 and rrr starts

Not worth flogging!
Owain Glyndwr is offline  
Old 22nd Nov 2013, 16:09
  #892 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Furthermore ...

10,000 ft/minute TAS at FL 350 ISA + 11°C is about 54 kt CAS
HazelNuts39 is offline  
Old 22nd Nov 2013, 16:25
  #893 (permalink)  
 
Join Date: Jun 2009
Location: florida
Age: 81
Posts: 1,608
Received 52 Likes on 15 Posts
Good to see some re-hash of a few things, especially the AoA issues.

As far as the pilot actions go, all we know is what the pilot did. Until we all join him in that hootch bar in the sky to get his version of the story, we shall never know.

OG made a good point many threads ago, and recently referenced. AoA "protections" depend upon mach bias once getting on up there WRT mach. Otherwise, the raw AoA is a very good indication of your state. Due to our mission requirements, and our relaxed static stability ( actually negative) until 0.95M, that bias was essential. Doze and I have talked about this and seem to agree that our implementation was necessary. So I return to the discarding of AoA once the dynamic pressure inputs to the FCS were deemed invalid. And BTW, on my emergency landing HUD tape you will note that all the displayed data went to zero when I reached 60 knots. Hmmmm. Of course, we were in "direct law" once we had WOW.

I have always been puzzled looking at the FCOM re: flight control laws, as it appears the system abandons many limits and such at the very first reversion mode. Roll control is one biggie, as 'bird points out with possible PIO for the first 30 seconds or so. And then there's the AoA protection, and sensed dynamic pressure be damned. Too easy to see if the vanes are cranking out good data. Our system had the traditional vanes ( conical sensors with slots), but we also had a "hemispherical" probe that had holes drilled all over it. That probe had its own heater and provided differential pressures to the "air data convertor", which then provided inputs to the FCS and an output for display and to our weapon/nav systems.

It is too easy to retain roll control laws other than the 33 degree limit and such, and simply command zero roll rate if the pilot is not moving the stick left or right. just look at the Thunderbird solo pilot's routines and note that the sucker rolls inverted and then the jet just stops rolling with no wing rock. It was easy, as all you had to do was relax pressure. Since there is no completely "direct law" except for the elevator/THS, this implementation has always puzzled me.

I wonder if the original test pilots ever went thru all the reversion laws where they could "fail" the air data or this data or...... I also wonder if they ever asked to fly in our little jet and see what we had using a similar test setup.

too long, but I had to get this off my chest.
gums is offline  
Old 22nd Nov 2013, 16:33
  #894 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi HazelNuts39,

Furthermore ...
"Time 2 h 11 min 58 - The vertical speed is around -15,300 ft/min."
That's around 80 kts CAS vertically - is that enough for you?
rudderrudderrat is offline  
Old 22nd Nov 2013, 16:42
  #895 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Hi rrrrr,

I thought we were discussing the moment the captain entered the cockpit and the stall warning stopped, which was at 02:11:44.

The stall warning operated again two times before 02:11:58.
HazelNuts39 is offline  
Old 22nd Nov 2013, 17:07
  #896 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi HazelNuts39,

2 h 11 min 45 End of “Stall, stall” warning + cricket.
2 h 11 min 53 SV : “Stall, stall” + incomplete cricket.
2 h 11 min 55 SV : “Stall, stall” + incomplete cricket.

If you had just walked onto the flight deck and heard a continuous stall warning and observed the other flight information - I bet you would have diagnosed a stall.
rudderrudderrat is offline  
Old 22nd Nov 2013, 23:52
  #897 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Owain Glyndwr
My point, not very clearly expressed, was really that 'forward speed' is an expression of Dozy's; I am sure that AB's engineers knew and know very well that it is the total air speed over the vanes that drives them not forward speed alone. [and so does Dozy I suspect, it was probably just loose wording]
Correct - a poor choice of words on my part. I'm a software guy, not an aero guy. My apologies.

The point I was trying to make stands, however. Whether you're building a system of an electro-mechanical, analogue electronic or digital electronic nature, excluding bad data is paramount, because you can always plan a system that will work predictably with an input missing or invalid, but if you allow bad data in, you'll get bad data - and feedback - out (GIGO principle).

@rudderrudderrat:

I said in this thread, and the other, that:

Originally Posted by DozyWannabe
I'd be very interested in seeing how all currently operating commercial airliners' sensory, control and warning systems behave when subjected to a flight profile that extreme - the problem is that no simulator can accurately replicate the conditions, and no test pilot in his right mind would risk flying that profile for real!
Because while it's all very well to hypothesise that the Captain might have diagnosed a stall earlier if the warning had still been sounding consistently, the fact is that we don't know how the systems on other aircraft behave when the aircraft is taken so far outside the expected flight envelope. To focus on that disregards the fact that the stall warning had been sounding consistently for around a minute without being acknowledged or acted upon by either of the flight crew present at the time. It also disregards the fact that the Captain was presented with unstable pitch tending towards nose-up, a rapidly unwinding altimeter and unreliable speed indications - would not these factors alone be enough to deduce a stall?

Last edited by DozyWannabe; 23rd Nov 2013 at 00:04.
DozyWannabe is offline  
Old 23rd Nov 2013, 01:30
  #898 (permalink)  
 
Join Date: Dec 2006
Location: The No Transgression Zone
Posts: 2,483
Received 5 Likes on 3 Posts
Pilots have overridden stick shakers and pushers before also
Pugilistic Animus is offline  
Old 23rd Nov 2013, 05:52
  #899 (permalink)  
 
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes on 0 Posts
@RRR:


Furthermore ...
"Time 2 h 11 min 58 - The vertical speed is around
-15,300 ft/min."
That's around 80 kts CAS vertically - is that enough for
you?
Cool,
Now the system has 'valid' V/S and 'valid' AoA.
Could it calculate a valid forward speed? FPA?

At that time the V/S indication was degraded, in excess of 10.000 ft/min the V/S is ADR only [amber boxed digital V/S display] i.s.o ADR + IR.

Last edited by A33Zab; 23rd Nov 2013 at 07:13.
A33Zab is offline  
Old 23rd Nov 2013, 08:33
  #900 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi DW,
but if you allow bad data in, you'll get bad data - and feedback - out (GIGO principle).
If someone could please explain how an A330 could ever have a CAS<60kts (at 1 g) whilst airborne - then I'll give up on this point.
If you are feeding "bad data" (intermittent / silent stall warning) onto the flight deck - then you'll probably get bad diagnosis out.

To focus on that disregards the fact that the stall warning had been sounding consistently for around a minute without being acknowledged or acted upon by either of the flight crew present at the time.
Please view
Audio clues tend to get filtered out by the brain during high work load. It is much better to use tactile (vibration / stick shaker).

It also disregards the fact that the Captain was presented with unstable pitch tending towards nose-up, a rapidly unwinding altimeter and unreliable speed indications - would not these factors alone be enough to deduce a stall?
Page 76 Interim report 3:
"In the absence of relevant information from the copilots, reading the information available on the screens (pitch attitude, roll, thrust, vertical speed, altitude, etc…) was not sufficient in itself for the Captain to become rapidly aware of the airplane’s situation. He did not then ask questions that could have helped him to understand the sequence of events.
The stall warning lasted 54 seconds continuously, during which time neither of the copilots made any reference to it. It is likely that the Captain heard this warning a few moments before coming back into the cockpit, but it is also likely that the multiple starts and stops added to the confusion and disturbed his diagnosis of the situation."

Last edited by rudderrudderrat; 23rd Nov 2013 at 11:30. Reason: extra text
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.