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

How safe is (airbus) fly by wire? Airbus A330/340 and A320 family emergency AD

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

How safe is (airbus) fly by wire? Airbus A330/340 and A320 family emergency AD

Old 21st Dec 2012, 16:46
  #41 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
Can the F/O action this selection strapped in?
BOAC is offline  
Old 21st Dec 2012, 18:00
  #42 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
How very tall is he ... ?
CONF iture is offline  
Old 21st Dec 2012, 20:56
  #43 (permalink)  
 
Join Date: Dec 2012
Location: Ramona, CA
Age: 66
Posts: 38
Likes: 0
Received 0 Likes on 0 Posts
How would you call a gradual and continouos nose-down input that cannot be overrided by the sidestick? Half a fist full?
When above the AOA at which alpha Prot activates, the sidestick becomes an angle-of-attack demand input. Thus, at neutral the sidestick commands the AOA equal to the onset of alpha prot. At full back, it commands alpha max (just shy of stall AOA).

The scenario at hand has the AOA vanes freezing in a fixed position and the Mach number increasing such that that AOA now moves within the alpha prot range with the increase in Mach number . (recall that the stall angle of attack decreases with Mach numbers > 0.3).

So, as the Mach number is increased to the initial alpha Prot value there is no change, except the stripped alpha prot band is now visible on the airspeed scale - where you would not normally expect to see it.
As the Mach number increases, the fixed AOA is now closer to the critical AOA, and the sidestick must be pulled slightly farther back to maintain this same AOA which is now farther within the AOA range. (So, yes, perhaps "half a fist full")
If the airplane continues to accelerate, the problem continues until such point that the frozen AOA equals Alpha max at the higher Mach no. and now despite full back stick input the airplane attempts to reduce the AOA by pitch down. (which will of course, increase the speed /Mach and make speed/Mach reduction - even with idle power - shall we say "difficult."

Thus, the solution is to not accelerate (increase Mach) when this happens (or to decrease Mach to below 0.3), which in and of itself, would probably be enough. Of course to take the AOA out of the control picture is a more complete immediate solution, and one that would be required once you got to the point of having full back sidestick. Switching off two ADRs, leaves the computers with one source of AOA data (not enough to rely on) thus removing them from the flight control protection picture, and solving the erroneous AOA protection problem.
Bpalmer is offline  
Old 21st Dec 2012, 20:56
  #44 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by ok
Since both SS are relatively useless at this point and the TL's don't move anyway,
- and in some airlines, both pilots are 'relatively useless' too.
BOAC is offline  
Old 22nd Dec 2012, 03:09
  #45 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by BOAC
- and in some airlines, both pilots are 'relatively useless' too.
With all due respect, it appears that the crew in this incident were perfectly able to handle the problem and recover the aircraft with no forewarning of the failure mode. Given that, is it fair to say that we should hold off on getting on our hobby horses about automation, pilot skills etc. before we have more facts to hand?
DozyWannabe is offline  
Old 22nd Dec 2012, 03:31
  #46 (permalink)  
 
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes on 0 Posts
Snoop

There is something wrong in the absence of feeling restitution by the feedback of some FBW flight laws. It was already stigmatized by gums' experience in flight.

In the direct loop you have the human feeling/sensors. From the feedback you get no feeling coming to the comparator.

In automation/system design it means there is a faulty transmission of information, with a resulting faulty OBSERVABILITY / CONTROLLABILITY of your closed loop system.

We know that OBSER ABILITY / CONTROLLABILITY are the minimum design qualities for a dynamic system to work.
RH
roulishollandais is offline  
Old 22nd Dec 2012, 04:02
  #47 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by roulishollandais
There is something wrong in the absence of feeling restitution by the feedback of some FBW flight laws. It was already stigmatized by gums' experience in flight.
That's a matter of conjecture and debate.

Those who favour the "traditionalist" line that all aircraft should be flown in the manner of a Cessna will probably agree, but others - of which there are a growing number - do not.

The former group are undoubtedly more vocal, but scratching the surface among many of them reveals a tangle of political motivations and technical misunderstandings that in truth serve only to cloud the real issues and hinder discussion.

The fact that in the years since their introduction, FBW Airbus models have maintained a safety record that equals or betters types with the traditional layout, and we have examples of safe returns after engine failures, successful deadstick landings (one of which was on water) and runway overruns in which every person on board survived. At the very least, the safety records of those types imply that the worst fears of the sceptics were somewhat overstated and never realised.
DozyWannabe is offline  
Old 22nd Dec 2012, 04:48
  #48 (permalink)  
 
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes on 0 Posts
Snoop spelling

Dozy
Analyzing which information comes to the comparator of closed loop system is not "traditionalist", or cessna " ,or "political", or "ground scratching", etc!
Your imagination has no limit ,nobody is attacking you, hiden behind equations

Last edited by roulishollandais; 22nd Dec 2012 at 07:06.
roulishollandais is offline  
Old 22nd Dec 2012, 05:16
  #49 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
As someone who speaks practically no French, I can't fault you for trying to express yourself in a language that is not your own, but I must admit I'm having real trouble understanding you - my apologies.

All I can say is that I read the article you pointed me at regarding "observability", and if I recall correctly I do not think it means what you think it means. It seems to me that you're making an argument for variable tactile feedback as a necessity (which is an assertion not supported by evidence as there has not been a crash caused by the lack of it in 24 years). Furthermore if that is what you're trying to say, I can't see the relevance to the subject under discussion. Forgive me if I'm reading you wrong.

Last edited by DozyWannabe; 22nd Dec 2012 at 05:17.
DozyWannabe is offline  
Old 22nd Dec 2012, 06:58
  #50 (permalink)  
 
Join Date: Jun 2011
Location: france
Posts: 760
Likes: 0
Received 0 Likes on 0 Posts
Snoop

Shortened, "observability" ("observabilite " in french system theory language) means that you have enough SENSORS. "Controllability" ("gouvernabilite" in french system theory language) means you have enough and coherent ACTUATORS. Put that in a closed loop with a feedback coming to the comparateur, and "pilot " the multivariable difference (we say "piloter l 'erreur " in french)).
It is the center of automation, regulation, robotik, system theory,call it as you want.
RH
roulishollandais is offline  
Old 22nd Dec 2012, 14:10
  #51 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Bpalmer
As the Mach number increases, the fixed AOA is now closer to the critical AOA, and the sidestick must be pulled slightly farther back to maintain this same AOA which is now farther within the AOA range.
If engaged, that's also the time for the AP to disconnect.
Now, depending on the angle at which the AoA probes froze, there is even a possibility for the A/THR to trigger TOGA With no way around to cancel it except than reaching those 2 ADR switches ... What a potential MESS !
CONF iture is offline  
Old 22nd Dec 2012, 15:33
  #52 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Chances of that are pretty remote though, and even in the worst case scenario you'd have a lot of time to sort things out before that happened.

Remember this is just a temporary workaround that applies to only a tiny fraction of the fleet, no need to go clutching at our pearls just yet...
DozyWannabe is offline  
Old 23rd Dec 2012, 02:15
  #53 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by DOZY
Chances of that are pretty remote though, and even in the worst case scenario you'd have a lot of time to sort things out before that happened.
Only a guy with absolutely no Experience being in the air in the front seat would dare making such comment ...
CONF iture is offline  
Old 23rd Dec 2012, 02:25
  #54 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
OK, so why is there no mention of Alpha Floor engaging in the incident concerned? Theoretically speaking it will eventually engage, certainly - but only if the Mach number increases sufficiently for it to do so, or if an attempt is made to pull up that is sufficient to trigger it.
DozyWannabe is offline  
Old 23rd Dec 2012, 03:09
  #55 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi DozyWannabe,
or if an attempt is made to pull up that is sufficient to trigger it.
If the AOAs are frozen in their present position, then pulling up won't have any effect on the AOAs.

I still don't understand why a simple "big red button" was never designed to turn HAL off and restore control to the pilots. Turning off 2 ADRs, with the subsequent loss of one PFD's Altimeter and airspeed information, does not seem to be an ideal way to accomplish that simple effect.
rudderrudderrat is offline  
Old 23rd Dec 2012, 15:07
  #56 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by rudderrudderrat
I still don't understand why a simple "big red button" was never designed to turn HAL off and restore control to the pilots. Turning off 2 ADRs, with the subsequent loss of one PFD's Altimeter and airspeed information, does not seem to be an ideal way to accomplish that simple effect.
Because dropping to Direct Law means that pilots who aren't used to manually trimming the thing outside of a simulator will have to very quickly get used to doing so in what is likely to already be a high-pressure situation.

Agreed that the workaround procedure is not ideal, but as I said before, and as the EAD clearly states, this workaround is temporary and only applies to a relatively small number of airframes.
DozyWannabe is offline  
Old 23rd Dec 2012, 15:54
  #57 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi DozyWannabe,

The same pilot will have to complete the approach and landing in Direct Law anyway once the Landing Gear is lowered.
If you are so concerned about a pilot's ability to fly this aircraft in Direct Law, then don't you think that perhaps it should have been designed so that the average pilot could do so easily?

this workaround is temporary and only applies to a relatively small number of airframes.
I am told there are 690 aircraft affected.
rudderrudderrat is offline  
Old 23rd Dec 2012, 15:58
  #58 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by DOZY
Because dropping to Direct Law means that pilots who aren't used to manually trimming the thing outside of a simulator will have to very quickly get used to doing so in what is likely to already be a high-pressure situation.
Big deal !
If manually trimming became one of those scary things ... (?), maintain the auto-trim but keep the bloody protections away on a single switch activation.

Manual trim or mad protection ? ... Maybe your choice is difficult, but it is not to anyone flying those things.
CONF iture is offline  
Old 23rd Dec 2012, 16:10
  #59 (permalink)  
 
Join Date: Jan 2006
Location: here, there, everywhere
Posts: 279
Likes: 0
Received 0 Likes on 0 Posts
I've read the EAD/OEB and the following passage puzzled me:

"In stabilized flight path (out of maneuvers involving an increase in load factor such as turns or
pitch variations), the Alpha Max strip (red) on the speed scale of the PFD can hide completely
the Alpha Prot strip (black and amber)"


Are Valphaprot and Valpha calculated from different parameters? The passage suggests that Valphamax is AoA and Mach dependant, while Valphaprot isn't (otherwise it would move up too?). Correct?

Then, this passage:
"With the AutoPilot (AP) engaged, and with the speed brakes in the retracted position, during
maneuvers involving an increase in load factor such as turns or pitch variations, the Alpha Prot
strip (black and amber) can move rapidly by more than 30 kt"


Why would that happen? does it mean jamming and then sudden movement of AoA vanes? If the AoA vane remained fixed during a maneouver, why would Alpha Prot move up?
Stuck_in_an_ATR is offline  
Old 24th Dec 2012, 00:46
  #60 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Alpha Max and Alpha Prot are dependent of the weight, config and Mach.
I would think the quoted extract : In stabilized flight path (out of maneuvers involving an increase in load factor such as turns or pitch variations), the Alpha Max strip (red) on the speed scale of the PFD can hide completely the Alpha Prot strip (black and amber) is related to the flight at higher altitude where Alpha Max is so close to Alpha Prot that the only visible strip on the PFD is the red one.

Regarding the second extract : With the AutoPilot (AP) engaged, and with the speed brakes in the retracted position, during maneuvers involving an increase in load factor such as turns or pitch variations, the Alpha Prot strip (black and amber) can move rapidly by more than 30 kt, it just make sense that V Alpha Prot increases under load factor, just as V Alpha Stall does. The question, as you put it yourself earlier, could be, why V Alpha Max does not at the same rate ?

Originally Posted by OK465
For the 330/340, IIRC alpha floor won't activate above 0.53 mach.
This would be a Question for HazelNuts39, at which angle the AoA probes should freeze for Alpha Floor to activate before M.53 is reached ?
CONF iture 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.