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, 02:00
  #1781 (permalink)  
Thread Starter
 
Join Date: Jun 2009
Location: florida
Age: 81
Posts: 1,608
Received 52 Likes on 15 Posts
What Turbine said

Salute!

Amen, Turbine.

To our Brazilian friend, I echo Turbine's feelings and goals, as all of us here do.
gums is offline  
Old 19th May 2011, 03:39
  #1782 (permalink)  
 
Join Date: Jun 2010
Location: SBA
Age: 56
Posts: 34
Likes: 0
Received 0 Likes on 0 Posts
Cargoflyer

+1
Khashoggi is offline  
Old 19th May 2011, 03:58
  #1783 (permalink)  
 
Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,810
Likes: 0
Received 0 Likes on 0 Posts
Cool

Hi,

Air France crash probe points to pilot error - NYPOST.com

The original article (Wallstreet Journal) offer a better analysis....
Airbus: No Major Malfunctions in Air France Crash - WSJ.com
jcjeant is offline  
Old 19th May 2011, 04:09
  #1784 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
Human touch

NandoCarioca,

valuable and trustworthy
A multidisciplinary group of aviation professionals and enthusiasts are here trying to contribute. The Synergy indeed is very high.

all information, regardless of its nature, should be posted and thoroughly discussed.
Most of pertinent postings are being commented and your presence is an special one. With a human touch, the ultimate reason of our interest in air safety discussing the issues each one in his field consider important.

trust me itīs been most helpful to all of us!
This is the direct result of the synergy here and your motivation.
this forum is a bliss among the anguish developed
I am sure the "search of the truth" is the same in the "investigation team" and may be this forum helped to show the aviation professionals are highly motivated to do the best flying or not. You could observe how many experienced retired professionals are answering questions sharing their knowledge.

Thank you all!
You are welcome.

And, at the time, was devastated to loose the opportunity to spend her vacations in PARIS.
For sure this facts bring to her something important for her mind.

Thank you for the human touch to a group of technicians. This make us remember our ultimate goal.
RR_NDB is offline  
Old 19th May 2011, 04:24
  #1785 (permalink)  
 
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 81
Posts: 1,461
Likes: 0
Received 0 Likes on 0 Posts
Air France crash probe points to pilot error
As we know here, the newspaper headline above results from reading between the lines and cannot yet be relied upon.

Let me pose a hypothetical question to those here.

Suppose it was true that the pilots blundered into a storm for whatever reason, and suppose that the aircraft load limits were not exceeded by the storm encounter.
Now suppose that some aspect of the flight control system's ability to control the aircraft while in the storm was exceeded and that led to the loss of control and crash.

Yes there are several suppositions here but as of this moment, these suppositions seem to be still potentially viable. I did post such a flight control possibility several months ago but did not pose the following question at that time.

Would you give the aircraft's control design a pass or a fail if that was the link that "broke" first in the above situation?
Machinbird is offline  
Old 19th May 2011, 05:12
  #1786 (permalink)  
 
Join Date: Jun 2010
Location: SBA
Age: 56
Posts: 34
Likes: 0
Received 0 Likes on 0 Posts
Machinbird

Fail.

If after flying a serviceable aircraft into a cb, and it does not ultimately exceed the airframe load limits, but does confuse the flight control system, that cannot be manually, at the pilots discretion, completely disconnected, to the point that the flight control system commands certain maneuvers that exceed airframe load limits and/or irretrievably exceeds the flight envelope, then yes FAIL.

Excuse the run-on.
Khashoggi is offline  
Old 19th May 2011, 05:37
  #1787 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
Machinbird,

some aspect of the flight control system's ability to control the aircraft while in the storm was exceeded
"System ability", relates to Software (algorithms, etc) an the Hardware of a "relaxed stability a/c" (Computational capabilities, Redundancy, fuel transfer for CG, ctrl. surfaces, etc).

After a System degradation for any reason, you need to insert a third element in the System: The pilot(s).

Your question, as i understood is about System ability to cope with conditions before the degradation.*

Would you give the aircraft's control design a pass or a fail...?
Software and the algorithms are "evolving entities". The designers are always looking for improving. Trying to attain the "State of the art".

The in depth analysis of the FDR data together Airbus SAS designers could answer your question. Itīs a complex issue.

And the leaks we are observing may be are in preparation to deviate for a subjective discussion with this complexity.

In order to understand precisely your question could you remember the post #? I would like to confirm what i understood.

(*) After the degradation even CRM is "inside the loop". A highly unpredictable "System (a/c) performance"

Last edited by Jetdriver; 19th May 2011 at 14:54.
RR_NDB is offline  
Old 19th May 2011, 05:37
  #1788 (permalink)  
 
Join Date: Jul 2009
Location: Planet Earth
Posts: 91
Likes: 0
Received 0 Likes on 0 Posts
Would you give the aircraft's control design a pass or a fail if that was the link that "broke" first in the above situation?
I wondered about this very question in my last post. I think it's more complicated than a simple pass or fail answer. This is a question that deserves some nuanced thinking about the fundamental design philosophy with regards to automation.

Having said that, if you put a gun to my head, I'd say its a pass. The time to prevent this tragic accident was before entering the storm.
CogSim is offline  
Old 19th May 2011, 05:52
  #1789 (permalink)  
 
Join Date: Jun 2010
Location: SBA
Age: 56
Posts: 34
Likes: 0
Received 0 Likes on 0 Posts
If flying into a cb is verboten, which of course it is, but to the point that a loss of aircraft is likely because of a flight control system that doesn't appreciate severe turbulence then several things need to happen:

The wx radar has to be on the MEL and a diversion required if it fails.
Pilot training has to be ensured on the make & model of wx radar.

If it really is critical and will cause a loss, there should be some type of wx radar redundancy, especially for long haul aircraft.
Khashoggi is offline  
Old 19th May 2011, 06:24
  #1790 (permalink)  
 
Join Date: Jul 2009
Location: Planet Earth
Posts: 91
Likes: 0
Received 0 Likes on 0 Posts
(*) After the degradation even CRM is "inside the loop". A highly unpredictable "System (a/c) performance"
If I understand this statement as meaning that human in the loop leads to unpredictability, I agree with you. However, unpredictability doesn't always imply unreliability. It may be possible to design more reliable systems by having humans in the loop well before degradation or even continually. The challenge will be to come up with an interface.
CogSim is offline  
Old 19th May 2011, 06:25
  #1791 (permalink)  
 
Join Date: Nov 2006
Location: SoCalif
Posts: 896
Likes: 0
Received 0 Likes on 0 Posts
Once you are beyond basic handling skills, in normal operation you find that the system in the plane that demands the most interpretive skills is the conventional Wx radar.
Graybeard is offline  
Old 19th May 2011, 06:28
  #1792 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
This can maybe appears in the next BEA report ...

The people who wrote that article should be utterly ashamed of themselves. There was no "co-pilot" on the plane. All three were full up qualified pilots. Having a "co-pilot" at the controls in no way "guarantees" that it will have problems. Securvol makes the Le Figaro article seem sensible by comparison.

There may have been only one person at the controls at the moment of upset. Supposing that person was one of the licensed flight crew he was a pilot, fully type qualified with a not inconsiderable amount of experience total and in type. Not enough information has been released to say more.

edit: I feel filthy after reading that.
JD-EE is offline  
Old 19th May 2011, 06:28
  #1793 (permalink)  
 
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 81
Posts: 1,461
Likes: 0
Received 0 Likes on 0 Posts
RR_NDB
The link to the prior post is here in the first section of this thread. http://www.pprune.org/tech-log/39510...ml#post6211506
Specifically the part regarding hydraulic supply saturation & rate limiting.
I would also like to correct a statement I made there, in that the maximum control phase loss resulting from rate limiting appears to be 90 degrees of phase loss.*
HN39 has since posted that a pitch up during cruise at altitude would not break the airframe (at least under positive g) since the wing would stall first.
Note* If the control signal is not perfectly in phase with the aircraft, it may still be possible to generate an out of phase control surface output.

I posed the question specifically because there may need to be a software change to extend the capability of the aircraft to handle turbulence (provided hydraulic saturation/rate limiting is proven). At this point, without hard information, all of this is still speculation as it relates to AF447.
My personal opinion is that if the control system cannot realize the full potential of the rest of the airframe, then it is a fail.
Machinbird is offline  
Old 19th May 2011, 06:38
  #1794 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
A FAIL case?

Machinbird,

The "Something Wicked" F117 (Vega 31) shot down during the Kosovo war can be considered as a result of a "control design fail"?

The intensity of the missile proximity fuse detonation made the system fail.

And whatīs the threshold?

Whatīs the threshold of a WX condition to an a/c?

SThis discussion has another problem: The System faced an UAS (like other 37 cases since 2003). The System simply cannot afford loose some very basic inputs, like air speed.

We learned that Airbus SAS is working in (two?) solutions for this:

1) Use other references (during short period) to avoid System instabilities.
2) Use other methods for air speed measurement.
x)

Additional comment: "Complex Systems" force pilots to respect "Complex WX" using "simple methods": A diligently operated WX radar (scanning manually, at different antenna elevations) to better "characterize" what lies ahead.

PS2

On the evolution of a System look the page 13 of the A330 mfr at 2004:

"A total approach to dependability"

Last edited by Jetdriver; 19th May 2011 at 14:50.
RR_NDB is offline  
Old 19th May 2011, 06:49
  #1795 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by RR_NDB
ITHO if they washed and dried it before powering up very probably we saved the portable VHF.
Right on the mark. It is also a way to clean up very dirty electronics, except for all kinds of disk drives that contain moving elements.

Pop out the clock battery. Wash it down well. Ideally use deionized water. But tap water can work. Then bake it out for a while at a modest temperature (120F or so.) Once it is completely dry power it up and go. Completely dry may take quite awhile at elevated temperatures even in very dry climates.

My sell phone survived a dunking in the mud via that kind of treatment.
JD-EE is offline  
Old 19th May 2011, 06:57
  #1796 (permalink)  
 
Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,810
Likes: 0
Received 0 Likes on 0 Posts
Cool

Hi,

More news !

Black hole

Google Vertaling

Original link:
L?AF 447 n?a pas foncé dans le trou noir - Europe1.fr - France
jcjeant is offline  
Old 19th May 2011, 07:00
  #1797 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
Unpredictability

I put unpredictable because most of the design effort was put to avoid the extremes.

After the "threshold" there are a lot of extra issues. And the interface is crucial.

One thing is clear: The System manager (PF) ideally should be capable to "avoid" (WX radar use, Comm. resources use, etc.) and to properly "administrate the crisis" after a System degradation.

But the optimization of the advanced designs "comes at a price" when unexpected (or unexpectable") situations arise.

The NYU risk Engineering professor introduced the "Black Swan" concept about "the highly improbable" events. I posted earlier a mention on him.

May be this case after BEA analysis may fit.
RR_NDB is offline  
Old 19th May 2011, 07:02
  #1798 (permalink)  
 
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 81
Posts: 1,461
Likes: 0
Received 0 Likes on 0 Posts
RR_NDB
The "Something Wicked" F117 (Vega 31) shot down during the Kosovo war can be considered as a result of a "control design fail"?

The intensity of the missile proximity fuse detonation made the system fail.

And whatīs the threshold?
OK, cute, but not relevant to AF447.

Whatīs the threshold of a WX condition to an a/c?
Not sure quite what you are implying. Why have an aircraft that could handle the worst weather it might encounter enroute except the flight control system can't hack it (but could if it was told how to)?
The System faced an UAS (like other 37 cases since 2003). The System simply cannot afford loose some very basic inputs, like air speed.
UAS is an assumed causal factor by BEA but not yet proven. Depending on what came first in the loss of AF447, it might still be the result of a departure, not the cause. But I will absolutely agree that they need a better system to provide airspeed. The control design is too dependent on it as presently structured.
Machinbird is offline  
Old 19th May 2011, 07:08
  #1799 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
Lonewolf_50, are you familiar with how a l@ser ring inertial reference works? Your remarks suggest you are unfamiliar with it and are extrapolating from physical gyroscopes. I'm sitting here puzzling over many aspects of your comment.

First, how do you cause a l@ser ring Intertial Reference Unit, IRU, to tumble. There is no moving part to tumble. My understanding is that the three rings are bolted down relative to the plane. The planes changes heading, pitch, and yaw are digitally processed to get the actual heading, pitch, and yaw relative to their state when the IRUs were initialized. Then you play with accelerometers (and to a degree the laser signals) to determine your position from integrating your direction of movement relative to the plane and its absolute heading. Then, since its most useful to know "up" relative to the surface under you rather than the tarmac on the airport you left the absolute reading is translated in coordinates to match your current position.

Once calibration is lost you probably cannot regain it in a trustworthy manner until you are on the ground.

And where in the BEA reports does it declare the IRUs themselves all went
unreliable at the same time. This exhausts my imagination trying to figure out how thus could happen short of a total power cut.
JD-EE is offline  
Old 19th May 2011, 07:12
  #1800 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
RR_NDB asks why not better fidelity on the CVR for the headset microphones.

I'd suggest that by the time you get through the first stages of audio processing past the microphone the signal is already bandwidth limited to what the CVR records. It's all that's needed to figure out what they said and more.

They're not planning to feed the signals to a top forties jock on FM.
JD-EE 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.