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

AF 447 Search to resume (part2)

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

AF 447 Search to resume (part2)

Old 19th May 2011, 14:35
  #1821 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 82
Posts: 1,688
DJ77;

Thanks for the added precision. Does that prevent the pilot from flying "the airplane according to aerodynamics and piloting experience"?
HazelNuts39 is offline  
Old 19th May 2011, 14:37
  #1822 (permalink)  

Sun worshipper
 
Join Date: Dec 2001
Location: Paris
Posts: 494
And we're back again fighting amongst ourselves for the quality of the Airbus construction / philosophy / floght control laws... etc...
Desitter,
First, let me apologize for my frustrated outburst of the other day - I was perhaps one beer beyond inhibition
Are you in the same state of mind ?
We all know how good a pilot you were and how well you could cope with airplanes that were built for virile, hairy pilots.
Forget it. Just try and understand how aviation has evolved and how much safer it is now.
There are hundreds of posts on the particulars of modern FBW airplanes and I suggest you read them before you'll make a complete cretin of yourself ( which hasn't happened yet)

DJ77
Note the ambiguity: nothing is said if Alt law 2 is due ADR faults or disagree conditions.
I'm going to give you *a benefice du doute* as I don't know wheteher your trolling or acting in good faith.
Your reference is only on the pitch axis which happens to have the same law in pitch whether you're in ALT 1 or 2. The Lateral modes are something else entirely as they are in Norm on ALT1 and a combination of DIR and ALT in roll and yaw.
From that original mistake, your post is dead wrong and you'll have to try again.

BOAC,
I read above that there is a failure mode which requires an inordinate amount of careful button pushing on an overhead panel to pass control to the pilot - no simple 'press one big button (easily to hand) and I will fly this aircraft'- or have I been mis-informed?
Don't know what you're talking about.
As far as this accident is concerned, and supposing that there was no return to normal on these systems, the Autopilot disengaged for a *Double ADR Fault*, the whole system reverted back to the pilots with an ALT2 FC law, which basically implies one has full control of the aircraft without protections.

Last edited by Lemurian; 19th May 2011 at 15:24.
Lemurian is offline  
Old 19th May 2011, 14:40
  #1823 (permalink)  
 
Join Date: Feb 2008
Location: Toulouse area, France
Age: 91
Posts: 439
"Staying ahead of the aircraft"

Cogsim's point about staying ahead of the aircraft is important for all "drivers airframe" - in the long-gone days when I was training to become one, I was firmly told that you din't just "stay" ahead, but had to be ahead of it - and being ahead started right at the start of briefing for the sortie, or even earlier if, say, there was an early morning weather briefing and your detail was for later in the day.
That kind of "being aheadedness" also applies to sailors, and even drivers ... (in an ideal world, of course).

This is just to amplify Cogsim's point and NOT to imply any reference to AF procedures, or what the crew did in this instance.
Jig Peter is offline  
Old 19th May 2011, 14:43
  #1824 (permalink)  
 
Join Date: Aug 2007
Location: London, New York, Paris, Moscow.
Posts: 3,631
Only one thing is certain at the moment, that is "stories" sell news-sheet.

Time to disengage from the media feeding frenzy and await the factual report from the BEA.
glad rag is offline  
Old 19th May 2011, 14:59
  #1825 (permalink)  
 
Join Date: Jun 2009
Location: VA, USA
Age: 56
Posts: 561
To return to an earlier point - the training capability of current simulators "beyond the normal flight envelope".

This is an area of current review within the flight simulation and training world.

Firstly I should note that virtually every current FFS in use is only qualified within the bounds of the airframe manufacturer provided data package, which will cover all "normal" flight envelope bounds, however MOST will not behave realistically beyond, and will likely use simple extrapolations once you err beyond 'normal'.

However it is obviously possible to model the behavior beyond the test flight data, and those with better aerodynamics understanding may be able to comment better, but the RAeS currently has an international working group titled ICATEE working on recommendations for the next generation of simulators to provide upset and extended flight envelope training.

For those interested here is the mission statement/plan for ICATEE as originally defined in 2009:

INTERNATIONAL COMMITTEE FOR AVIATION TRAINING IN EXTENDED ENVELOPES (ICATEE).

On 3 and 4 June 2009, the Flight Simulation Group (FSG) of the Royal Aeronautical Society organized and ran a Conference entitled Flight Simulation - Towards the Edge of the Envelope. The Conference subject was related to a growing need to address the aviation safety issues arising from loss of control in flight through better training and simulation beyond what is currently covered in flight and in ground-based flight simulators.

The Conference identified the need to improve aircraft upset training, the shortcomings in basic education and readiness of commercial pilots in reacting to upsets, and the technical challenges. At the conclusion of the Conference, the RAeS FSG was requested by the delegates to form a Committee to explore the formation of an International Working Group (IWG) to identify follow-on steps and to invite participation from other interested parties.

The FSG has considered this request and subject to Society approval plans to proceed with the leadership of this Working Group which we have titled the International Committee for Aviation Training in Extended Envelopes (ICATEE).

The ICATEE objectives can be stated as follows:

-To be a body independent of national, corporate or factional interests, amalgamating information on the subject matter
-To identify the problem and recommend prioritization for remedial action
-To identify and help qualify current processes and methodologies
-To collect data on enhanced envelope maneuvers
-To interact with research concerning improved training in extended flight envelopes
-To identify and recommend best practices
-To identify future related research activities
-To support the implementation of regulatory guidelines or updates for enhanced training

The main activities of the Working Group will be:
- to review current practice in extended envelope training
- identify the main shortcomings
- identify the data and training media requirements and further research required
- recommend the areas that need immediate improvement in flight training
- examine essential and desirable training elements
- suggest how standards can be established.

The FSG has established an ICATEE steering group of FSG members consisting of Sunjoo Advani (chair), Gordon Woolley (co-chair) and Peter Tharp (co-chair). This group will set up and confirm the plan of the Working Group, assign tasks, request the support of the participants to fulfill these responsibilities, establish the time lines,and report the findings outside the group. It will also prepare the Committee’s Constitution and Terms of Reference.
The RAeS Flight Operations Group has been invited to participate and the Human Factors and Air Transport Groups will also be kept updated with progress.
Participants in the group will be expected to finance their own involvement. No funding will be requested or required from the Society.

It is the intention to present the findings of the ICATEE at an RAeS Conference in the first half of 2011.

The first form of dissemination of the information gained within the Working Group will be through the Royal Aeronautical Society FSG meetings and conferences. It is proposed that several channels be established in order to disseminate the information from the Working Group:

-a web-based forum through the RaeS FSG website
-a limited number of meetings
-presentation at the appropriate conferences (e.g. FSG, AIAA, FSEMC, WATS)
-industry white papers to feed back key findings to the stakeholders.
-inputs to the International Committee for FSTD Qualification (ICFQ)

The steering group met on 23 July, and is planning the following actions:

1. Circulate the Master Plan document and invite participation in the Working Group
2. Issue Constitution and Terms of Reference, and obtain FSG approval
3. Establish the Working Group on the basis of responses to the invitation
4. Request the Working Group at its initial meeting to:
a. Review its objectives
b. Identify Work Packages, leaders and assistant leaders. See Appendix 1 for possible Work Packages.
c. Define meeting targets
d. Agree outputs and timetable
e. Set methodology for future meetings and discussions
f. Set up a reporting procedure to the Steering Group
See the RAeS FSG website for the latest developments.

- GY
GarageYears is offline  
Old 19th May 2011, 15:07
  #1826 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 62
Posts: 5,678
JD-EE

Vaguely, I should do a bit more homework to brush up on the basics. "Tumble" is the physical analogue (from old gyroscopes) I am trying to use to describe the function becoming unreliable, based perhaps on accelerometers going wrong, but I won't comment further as I think the [email protected] ring gyro has been adapted due in part to its higher reliability. (Fault tolerance).

As to lost calibration, causes of that would, it seem, remain in the electrical, rather than physical motion, domain. Do I understand that correctly?

I'll stop with the wandering off into partial panel wilderness. Sorry, BOAC, for bringing that up again. In the process, I've learned a bit, so maybe not a total loss. Where I think I wandered off was in pondering an ADIRU fault consideration. We have tended to discuss that more on the Air Data side (based on pitot tube / probe issues), rather than IR faults (Inertial Reference) within that system, based on the ACARS messages.

FDR will doubtless be much clearer regarding what was doing which.

That said, there have been (in a general sense) ADIRU faults in the past the pilots had to trouble shoot due to impact on their attitude reference display. The summary I just reviewed again shows crews to had sort them out or handle them via (IIRC a 777 incident) by degrading Auto Pilot funcitons ... perhaps while being in a bit more benign flight environment than T Storm penetration?

Thanks to OK465 for the sim training and unusual attitudes points. Appreciated.

Last edited by Lonewolf_50; 19th May 2011 at 16:09.
Lonewolf_50 is offline  
Old 19th May 2011, 15:30
  #1827 (permalink)  
 
Join Date: Jun 2009
Location: France
Age: 74
Posts: 193
Lemurian;

You must have an A330 FCOM. Have a look at 1.27.30 and tell us exactly what you understand about pitch control in ALT1 or ALT2, especially the use of speed in these laws. I never mentionned lateral control
DJ77 is offline  
Old 19th May 2011, 15:38
  #1828 (permalink)  
 
Join Date: Jun 2009
Location: France
Age: 74
Posts: 193
HN39;
Thanks for the added precision. Does that prevent the pilot from flying "the airplane according to aerodynamics and piloting experience"?
Not as long as a correct or "default" airspeed is used by pitch control I suppose.
DJ77 is offline  
Old 19th May 2011, 16:05
  #1829 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,581
Originally Posted by Lemur
Don't know what you're talking about.
- nor do I, and it was, to add to the confusion with two threads, in post #293 in R&N from Icepack.
BOAC is offline  
Old 19th May 2011, 16:06
  #1830 (permalink)  
 
Join Date: Jun 2009
Location: Earth
Posts: 79
Flight Control Laws

Gentlemen,

if I may, I intend to shed some more light on this touchy matter of Flight Control Laws on the accident type.

Each Flight Control Primary Computer will perform its own computations, and decide its own Flight Law, based on available data. FCPCs (or PRIMs) manage Normal and Alternate laws, while FCSCs (or SECs) manage Direct law. The Master PRIM (normally PRIM1) then decides which Law gets used for the whole system.

Lemurian wrote :

the whole system reverted back to the pilots with an ALT2 FC law
This is unfortunately not accurate enough in our case. I urge you to reconsider the very particular case of FCPC n°2, after it lost connectivity with ADR1 following this failure :

02:11:55 EFCS1 X2,EFCS2X,,,,,,,FCPC2 (2CE2) / WRG:ADIRU1 BUS ADR1-2 TO FCPC2, HARD

This is definitely a wiring fault, where FCPC2 and ADR1 lose their connection.

Lemurian, could you be so kind as to demonstrate with certainty that FCPC2 reached indeed a state where it operated in Alternate 2 law ? I am not so sure. Normal law would be the preferred mode even with only two ADRs, as long as they produce similar values.

FCPC1 certainly did operate in ALTN2, as correctly demonstrated by BEA in its intermediate reports. It was Master PRIM until it failed along with its SEC1 partner :

02:13:45 F/CTL PRIM1 FAULT
02:13:51 F/CTL SEC1 FAULT

Then FCPC2 became Master of both elevator halves and THS.
Svarin is offline  
Old 19th May 2011, 16:08
  #1831 (permalink)  

Sun worshipper
 
Join Date: Dec 2001
Location: Paris
Posts: 494
DJ77,
I never mentionned lateral control
That's a pity because I can't see how you could derive anything on your flight controls state if you did not know the initial *triggers* of the degradation to ALT.
But of course, just mentioning the pitch control allows you to criticize in a very sleazy manner this architecture. So, when you wrote
... / ALT LAW 2 / Pitch control
Identical to ALT 1 law" Note the ambiguity: nothing is said if Alt law 2 is due ADR faults or disagree conditions.
,
you very conveniently forgot to mention that the conditions for ALT2 are, among others :DOUBLE ADR FAULT ...ADR DISAGREE...

Thank you, the doubt I had is lifted.
Lemurian is offline  
Old 19th May 2011, 17:09
  #1832 (permalink)  
 
Join Date: Jun 2009
Location: Paris
Posts: 691
Originally Posted by Svarin
Each Flight Control Primary Computer will perform its own computations, and decide its own Flight Law, based on available data. FCPCs (or PRIMs) manage Normal and Alternate laws, while FCSCs (or SECs) manage Direct law. The Master PRIM (normally PRIM1) then decides which Law gets used for the whole system.
Basically, you can fly it in Direct Law with only one SEC remaining (all 3 PRIMs and 1 SEC lost). This aircraft has been fully certified in Direct Law.

Originally Posted by Svarin
This is unfortunately not accurate enough in our case. I urge you to reconsider the very particular case of FCPC n°2, after it lost connectivity with ADR1 following this failure :
02:11:55 EFCS1 X2,EFCS2X,,,,,,,FCPC2 (2CE2) / WRG:ADIRU1 BUS ADR1-2 TO FCPC2, HARD
This is definitely a wiring fault, where FCPC2 and ADR1 lose their connection.
Not sure what you are saying is right. If it is a wiring trouble, it is on BUS 2 without affecting BUS 1. This is also too much reading as ADR1 was also very likely declared unreliable to the master (PRIM1).

Originally Posted by Svarin
Lemurian, could you be so kind as to demonstrate with certainty that FCPC2 reached indeed a state where it operated in Alternate 2 law ? I am not so sure. Normal law would be the preferred mode even with only two ADRs, as long as they produce similar values.
There is not a single doubt that ALT2 was triggered from the begining of the sequence.

Originally Posted by Svarin
FCPC1 certainly did operate in ALTN2, as correctly demonstrated by BEA in its intermediate reports. It was Master PRIM until it failed along with its SEC1 partner :
02:13:45 F/CTL PRIM1 FAULT
02:13:51 F/CTL SEC1 FAULT
This is called a reset, not a failure.
takata is offline  
Old 19th May 2011, 17:36
  #1833 (permalink)  
 
Join Date: Oct 2009
Location: berlin
Posts: 152
PJ2 However, in two Flight Crew Training Manuals examined, fuel transfer is mentioned. It is stated that the handling characteristics of FBW aircraft are independent of the CG in Normal and Alternate Laws and it is not necessary to command a FWD fuel transfer in the event of heavy turbulence in cruise.
HarryMann, Tubby Linton, PJ2 for my this "independent declaration" is one of the WHY´s to AB,

why AB means that FBW did not longer need more stability in turbulence ??? what has changed between 1988 and 1997 ???

neither the basic aerodynamic of the sky nor the basic concept of the wing (profile) is quite different between an A300-600R and an A330-200 (both with trim tank)
grity is offline  
Old 19th May 2011, 17:48
  #1834 (permalink)  
 
Join Date: Jan 2008
Location: Los Angeles
Posts: 168
Ref deSitter's #1812 - Despite technical b/g my basic SLF status causes me to enter this discussion among a large group of wise and honorable (and sometimes agitated) elephants with considerable trepidation, but would it not be useful for the Flight Control Computer to show some sweat? It knows - or can be taught - how close it is to turning the job back to mankind.
poorjohn is offline  
Old 19th May 2011, 18:03
  #1835 (permalink)  
bearfoil
Guest
 
Posts: n/a
In reading the pilots of this type, I understand AB to require autoflight in turbulence. Likewise, the lack of need to rearrange the fuel components when anticipating flight into turbulence. This last requirement, that Fuel is allowed in the Trim Tank in Turbulence, applies to ALTLAW 1 and ALTLAW2, but what about DIRECTLAW? With an aftward cg, do control movements that are handled by the autoflight present a challenge to pilots in DIRECT that they are not used to reacting to? In the AB manual, for instance, references are made to cruise. What is the AB direction if not in cruise, or manual flight of any description, for that matter?
 
Old 19th May 2011, 18:07
  #1836 (permalink)  
 
Join Date: Jan 2011
Location: Pennsylvania, USA
Posts: 130
Would someone be kind enough to reply with a general summary of where the investigation stands now? I tried reading back through the threads but it seems to ah...change direction quite often or something.
Plasmech is offline  
Old 19th May 2011, 18:10
  #1837 (permalink)  
 
Join Date: Jan 2008
Location: Blighty (Nth. Downs)
Age: 75
Posts: 2,085
THS Trim Tanks. CVRs & "Hot" Microphones.

Hi grity,
Just a reminder that the A300-600 series are of similar vintage to the A310. (They retain the A300 wing, if memory serves, unlike the super-critical A310 wing.) They have conventional flight controls, and (presumably) a traditional-size THS.

Tubby Linton will be able to explain more about the MAC figures involved. However, speaking generally about swept-wing, medium-to-long-range jet transports those without trim tanks the following may help. When pushing range towards the limit, the centre tank has a substantial amount of fuel at take-off. (You only put fuel in it if all the wing tanks are already full.) Because the centre-section is forward of the optimum CG, the aircraft is nose-heavy and inefficient until the centre tank is empty. So during the first few hours, the THS is in an inefficiently nose-up-trim position (i.e., having to generate more negative lift than later on). Having the THS trim tank full does not necessarily create an abnormally aft CG, but it can if the designers want it to.

EDIT (May 20)
The point I was trying to make above is that the A300-600R (and A310-300) trim tanks are unlikely to be used to create "relaxed stability": merely to offset the normal forward-CG problem that I have illustrated.
The A330/A340 are a different case, because of FBW. This is not to suggest, however, that they would be unflyable in Direct Law at cruising altitude. Current certification criteria would not permit such a regime, if my understanding is correct.

Garage Years,
Take the point about NR headsets, which I used from the early 1990s. Being short-haul, we had our headsets on nearly all the time. What I can tell you is that, when the cabin crew entered the cockpit, we had little difficulty hearing them and conversing. Admittedly, we often slid one earpiece off, but I'm not sure that was necessary. My understanding and experience was that it was steady "noise" that was almost eliminated: e.g., the 400Hz hum from the AC electrics, the airframe noise, and the engine noise.

I'd be surprised if "hot mikes" (for CVR recordings) were phased out on UK-registered aircraft as a result of noise-reduction headsets. Do we know if Air France CVRs use them?

Last edited by Chris Scott; 20th May 2011 at 12:43. Reason: See 3rd paragraph. Last para extended.
Chris Scott is offline  
Old 19th May 2011, 18:42
  #1838 (permalink)  
 
Join Date: Jun 2009
Location: France
Age: 74
Posts: 193
Svarin, thanks for your inputs. I was thinking that the ADR1 / FCPC2 wiring failure could have been just a transient. I find it is unclear how exactly pitch control works in ALT1 / ALT2. Whilst these reconfiguration laws can be triggered by 2 or 3 unreliable airspeed sources the FCOM says it uses speed. I offered that the FCPCs may use a “default” speed value or default gains in case of missing (or unreliable) airspeed data. Good enough for me.

My point is: an FCPC in charge of pitch control in a load factor demand law, which is the case in both ALT laws, apparently have to establish, one way or another, reasonable gain values. Now what happens if in a particular sequence of events, especially when the ADR data may be restored randomly and asynchronously relative to the computer’s sampling rate, this FCPC is led to use (perhaps momentarily) an erroneous speed value and then determine wrong gains? Is the airplane always flyable in such a case?

Of course I have no doubt that all the software was carefully studied and checked in all possible ways but sh*t happens. I agree this may be far-fetched and stand ready to be guillotined.
DJ77 is offline  
Old 19th May 2011, 18:59
  #1839 (permalink)  
 
Join Date: Jan 2008
Location: Atlanta, GA, USA
Posts: 349
Irrelevance

All this discussion of what the Airbus FMCs do is interesting, but irrelevant - when humans are confronted by a computer that is either not working, or working in a regime they are not intimately familiar with, they freeze - they go into brain lock. This happens with all kinds of systems, from text editors to banking systems, and I'm sure it happens with aircraft FMCs. If you are going to design airplanes that fly on the edge of control with the help of computers to constantly trim them, then you'd better make sure that the airplane still manually flies like an airplane at the drop of a hat, one that a pilot can control *instinctively*, despite his brain lock! This is a criticism neither of pilots, nor of computer systems - it's the truth of human/computer interaction. It will never matter how well designed are Airbus' or Boeing's FMCs - if they are not purposely designed to make the airplane act like an airplane under pilot input, then they are not only wrong-headed, they are dangerous!
deSitter is offline  
Old 19th May 2011, 19:03
  #1840 (permalink)  
 
Join Date: Mar 2001
Location: us
Posts: 694
plasmech,

The BEA is continuing to do a read-out of the two recorders. There have been no disclosures of the content of the CVR, yet. The BEA has said that based on the initial read of the FDR, that no major hardware or software problems were found, but that as the data is further analyzed, anomalies or problems requiring corrective action or steps may be identified.

The two bodies recovered in their seats have been flown to Paris, and the examining laboratory has announced that the bodies can be identified using DNA.

The Ile de Sein is headed back to the site of the plane; it had been in Dakar for a crew change. Quite likely it off-loaded in Dakar those parts of the plane, e.g., engine, cockpit seats, that had been retrieved, and these are probably being flown back to France. It is not clear what parts of the plane will be retrieved, possibly cowlings if these were not retrieved in the initial recovery of parts. I believe two judges in Paris will now determine whether any more bodies will be retrieved.
SaturnV is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

Copyright © 2021 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.