Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

Airbus prepares safety warnings following A321 incident

Wikiposts
Search
Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

Airbus prepares safety warnings following A321 incident

Thread Tools
 
Search this Thread
 
Old 22nd Nov 2010, 06:54
  #121 (permalink)  
 
Join Date: Aug 2005
Location: fairly close to the colonial capitol
Age: 55
Posts: 1,693
Likes: 0
Received 0 Likes on 0 Posts
"Cowboys d'Airbus" by Bernard Ziegler
Tex Johnson with a taste for the fromage et vin?
vapilot2004 is offline  
Old 22nd Nov 2010, 17:40
  #122 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Fargoo
How many reset on FAC1 have been done for your video sequence ?
Is it something like 0.9 degrees increment per reset ?
CONF iture is offline  
Old 22nd Nov 2010, 19:10
  #123 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
jcjeant:
He was a decorated military pilot before he got too cocky and caused the Alps accident. He has also not worked for AI since 1997 IIRC.

All I'm wondering is if anyone will ever realise "Le cheval que je flagellé, il est mort".
DozyWannabe is offline  
Old 22nd Nov 2010, 19:21
  #124 (permalink)  
 
Join Date: Mar 2010
Location: overthehillsandmountains
Posts: 140
Likes: 0
Received 0 Likes on 0 Posts
Ziegler Jnr

Indeed. BZ is now an elderly and somewhat frail gentleman.

Lay off please.
kwateow is offline  
Old 22nd Nov 2010, 21:17
  #125 (permalink)  
 
Join Date: Mar 2005
Location: England
Posts: 730
Likes: 0
Received 0 Likes on 0 Posts
Fargoo
How many reset on FAC1 have been done for your video sequence ?
Is it something like 0.9 degrees increment per reset ?
It's approximately 0.1 degrees per reset and always to the right , I was hoping you would hear the click of the switch each time but the camera mic wasn't sensitive enough over all the other background noise.
Fargoo is offline  
Old 8th Dec 2010, 20:58
  #126 (permalink)  
 
Join Date: Aug 2006
Location: London
Posts: 51
Likes: 0
Received 0 Likes on 0 Posts
Captain Crunch

Could you miss the mountains in a pull up with this type of g-limiting and bank-limiting A320 FBW sidestick?
No, I don't think you can. FBW will save the airframe from overload first, but ultimately hit the rocks as I understand it. It will reject the pilot's command to pull say seven g's. FADEC will slow spool the engines to save TBO but hit the trees as it did in Toulouse. Flying around on Alpha mode on the back side of the power curve is insane below ten thousand feet but simulator gods, test pilots and check airman seemed unconcerned about fostering such a dangerous habit when we first got the airplane.
I am honestly dumb-struck at your ignorance of your own subject.

Higher g will only offer advantage in the first few metres of a pull-up, the ability to pull max alpha will give more advantage in almost any conceivable GPWS scenario, something which is much, much easier in an alpha protected FBW type. Numerous studies have shown that such aircraft will massively outperform conventional types in GPWS pull ups.

If you pull seven g then you don't need to hit the mountain, you will have destroyed your aircraft all by yourself.

No-one to my knowledge has ever 'fostered the habit' of flying on the back of the curve at low level (and I have been flying this machine for 16 years). It is a facility that can be useful in certain unusual situations. Flying the aircraft at SOP speeds (O, F, S) keeps you right in the flat part of the curve. If you are operating on the back of the curve then it is your poor practice, not a defect of the aircraft, its makers or its operating philosophy.

FADEC does not slow spool the engines to save TBO. It does so to avoid surge/stall. All large fan engines are slow from idle with or without FADEC. FADEC just makes the acceleration safer and more reliable.
320 driver is offline  
Old 8th Dec 2010, 22:30
  #127 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by '320 driver'
If you pull seven g then you don't need to hit the mountain, you will have destroyed your aircraft all by yourself.
In given circumstances, 3G or even just 2.1 or 2.6 could be all it takes to avoid ground contact ... too bad the electronic will not approve the request.

Pardon my ignorance but why do you associate G meter and alpha max ?
CONF iture is offline  
Old 8th Dec 2010, 23:00
  #128 (permalink)  
 
Join Date: Aug 2006
Location: London
Posts: 51
Likes: 0
Received 0 Likes on 0 Posts
the advantage of pulling an extra 0.1 or even 1.0g is miniscule and only beneficial if the obstruction is within a few metres of the aircraft when the maneuvre is initiated or the A/C is at very high speed. Overwhelming majority of GPWS incident are at low/medium speed. Modern GPWS will provide hard warning well outside the envelope where g is likely to be more useful than alpha performance.

The advantage of being able to accurately pull max alpha is huge and much more likely to prove beneficial.

I don't think I associated g meter and alpha. The point of my post is that they are separate and one (the one that A320 allows you to utilise much more effectively) is much more useful than the other.
320 driver is offline  
Old 9th Dec 2010, 06:53
  #129 (permalink)  
 
Join Date: Jul 2007
Location: ***
Posts: 350
Likes: 0
Received 0 Likes on 0 Posts
To add:

It will reject the pilot's command to pull say seven g's.
where:
Vst = stall speedVs = stall speed of the aircraft in straight, level flightn = load factor
So for a stall speed of 100 kts you get

Vst = 100 * sqr7 = 100 * 2,65 = 265 kts

So yanking 7g will not only break the wings off of any airliner, you'll most likely stall it and hit that mountain.

As explained above by 320driver to maintain alpha max will give you the maximum of achievable lift over the entire escape manouver.

That is also the reason, why the bank angle is limited to 66 degrees - flying a level, 66deg bank turn will result in 2.5g => which is the max gforce for which an airliner has to be designed for.

Nic
Admiral346 is offline  
Old 9th Dec 2010, 13:12
  #130 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,201
Received 401 Likes on 248 Posts
The aircraft rolled to the left and adopted an approximately 10ş left-wing-low attitude, without any flight control input from the crew. The flight crew reported that the aircraft did not seem to respond as expected to their control inputs and shuddered and jolted repeatedly.
First, a thought: if you make a control input and the aircraft does not respond as commanded, you are by definition in "out of control" flight ... so technically, the crew in this incident had a short excursion into "out of control" flight.

Curious regarding reaction from the pilots to this uncommanded control input from the robot.

When I was flying fleet helicopters with complex AFCS, the typical immediate action response (NATOPS BOLD FACE etc) was that uncommanded flight control inputs required immediate disconnect of AFCS, (don't consult the checklist, do it!). IF SAS or Boost was giving you spurious input, (AFCS being off, and the heirarchy being what it was) dump them as well until spurious input ceased.

As I read this report, the crew's reaction was similar but different. It appears that it took them a while to work through the "what's it trying to do now" and I gather that the intermittent nature of the fault drove that.

Am I following correctly? (Nice to be at 36K and have the luxury of time to work through a problem).
Unhooked - I recall hearing about an incident with a A340 out of MRU where the AOA vane was damaged in the stand and went unnoticed or unreported. ... I think the fast thinking commander switched off some of the PRIM's & SEC's (primary and secondary flight computers) and managed to return in direct law.
Fly the plane ... and disable any systems interfering with your flying the plane. Important point on that is that you really have to know your systems to ensure you know what does what ...
wiley - Complex systems fail in complex and often unanticipated ways.
Which is part of the problem of the "magic airplane" design philosophy.
You create X problems when you use robots to solve Y other problems. The only defense the crew have is expert and in depth system knowledge, and a robust continuum of training.
Clandestino - ... that Airbus is crash-proof and can cope with less skilled pilot than standard is misperception, promulgation of which is the fault of Airbus propaganda department, not the fellows who designed the Airbi.

Woe to the airline whose training department takes this sales pitch to be true.

Compared with a couple of decades ago, modern pilot has replaced: pilot, flight engineer, radio operator and navigator. Creating better and easier to operate systems was meant to decrease workload. It succeeded yet the intention was not to make life in cockpit easier: it was to reduce workforce as much as possible. It is all fine and well as long our electronic little helpers do their work. When they pack up, you have two people doing the job that was historically split between 4 to 6
Good point. Made a similar observation a few days ago, how complex aircraft and reduce crew actually up the requirements and demands on the crew. I found when doing crew training that any two man crew can eventually reach task saturation. (I had an evil reputation as a sim instructor, I confess! I liked to find crews' limits with multiple system malfunctions / emergencies ... unscripted.)
stev - A lot of the incidents of recent times have come from lack of knowledge of what the system is doing. Much of what Airbus tried to eliminate in the making of the A320 has been taken care of in the major developments of teaching of the modern elements of CRM. Leaving it to the computer may not be ideal.
Rananim seems to have a sound philosophy on that one.
rananim - I believethe airbus peole were well-intentioned and produced a landmark machine but the over-confidence was sickening...I like it this way.Monitor me,warn me but give me control.always.
Next thing you know, one breaks up over new york and the pilot is blamed for using the rudder. A memo comes out telling me not to use the rudder if I can help it!
Which brings us back to the incident in question.
From the report, it seems the rudder was being commanded (by other than a pilot) to do something that it should not have been doing. I will guess that perhaps -- and perhaps not -- the limitations in rudder control defelction were being respected by the robot (or not???) even though the robot, thanks to some dirty trons, was moving the rudder when it shouldn't. Different logic/control loops?

That is my layman's way of trying to decipher which protections may or may not have remained when one (of many channels) in the automatic flight control system went wrong.

It looks like the crew approached this malfunction in a rational, unhurried, and professional manner and got the bird to where it needed to be: on the ground where the electron chasing magicians would investigate and remedy the fault.


My surmise is that the robot will find another way to challenge a cockpit crew in unexpected ways in the not to distant future.

May it not happen close to the ground.
Lonewolf_50 is offline  
Old 9th Dec 2010, 14:49
  #131 (permalink)  
 
Join Date: Aug 2006
Location: London
Posts: 51
Likes: 0
Received 0 Likes on 0 Posts
Lonewolf

in the AA New York accident the rudder was being solely manipulated by the pilot. It was pilot error that generated the rudder input. The system had nothing to do with it.
320 driver is offline  
Old 9th Dec 2010, 16:19
  #132 (permalink)  
 
Join Date: Apr 2000
Location: uk
Posts: 121
Likes: 0
Received 0 Likes on 0 Posts
A320Driver
At last the voice of common sense.

I am honestly dumb-struck at your ignorance of your own subject.

Higher g will only offer advantage in the first few metres of a pull-up, the ability to pull max alpha will give more advantage in almost any conceivable GPWS scenario, something which is much, much easier in an alpha protected FBW type. Numerous studies have shown that such aircraft will massively outperform conventional types in GPWS pull ups.

If you pull seven g then you don't need to hit the mountain, you will have destroyed your aircraft all by yourself.
screwdriver is offline  
Old 12th Dec 2010, 08:22
  #133 (permalink)  
 
Join Date: Mar 2009
Location: on the ragged edge
Posts: 80
Likes: 0
Received 0 Likes on 0 Posts
The Bus Cult Continues

Interesting post by 320 driver.

I believe it aptly demonstrate my point about the "blind faith" some A320 drivers exhibit toward a design that over-rides the pilots command to miss the rocks since it's programed to instead save the airframe. The F-16 community went through a similar overconfidence in it's FBW introduction: that the airplane could get you out of any problem, with predictable fatal results. We're agreed then, that the studies of GPWS escape maneuvers at low altitude come into play AFTER Controlled Flight Into Terrain is avoided with an aggressive pull up, not before, in the scenario I refer to. I'm talking about a visual impact with hills due to a gross navigational error down low, and not spotted until too late. O.K, 7 g's was a lethal exaggeration. How about 3.5 G's? (thanks for the correction CONF iture).

Me? I prefer to bend the wings and miss the rocks, as was done on several aircraft that made it back alive due to the ability to temporarily exceed the design limits. I am aware of McDonald Douglas and Boeing events which were classified as accidents since the airframe was severely damaged in over-g pull ups. In Good Aerospace engineering, failure limit loads are not the same as operational limits, and have something like a ten to thirty percent additional protection before airframe yield occurs. Additionally, if an airplane is designed correctly, it will progressively fail, shedding outer structure before the spars completely fail.

I admit that I am completely ignorant how a profession pilot would worry about saving his airframe if rocks are looming in the windshield as they apparently were in Pakistan recently. Who cares about the airframe in this scenario? Miss the rocks at all costs, then, if you're still alive, hope you can limp back to the runway. But not in FBW as I understand it. In FBW you are not in command. The HAL-9000 is in command:

"I'm sorry Dave, I'm afraid I can't give you four g's right now, it might bend the titanium spar". BOOM.

Let me try to rephrase the question: Due to rapidly rising topography, if it takes say 3.8 g's immediate pull up in two seconds time to miss a rock outcrop and your A320 is given this command with full back stick, will it follow it?

No, it won't as I understand it. It will respect the load limit algorithm and only give you something like less than 2.x g's. Initially, in this case, the G limiting subroutine will overide any available angle of attack considerations.

Right?

Congratulations! We have saved the airframe for two seconds, but lost a hundred souls and the airframe in three!

Now I feel this is a design oversight. When I use that term or the term software`-shortcoming it's extremely upsetting to Dozy and others, however, it's not surprising. Aviation history is replete with design oversights. Don't worry Dozy, We are not blaming programmers (CEO's and flight management get that distinction), and besides, the event we're discussing is extremely unlikely an occurrence. But if it happens and you need say 80 degrees bank to miss the rock cliffs, you better pray you're in a Boeing.

By the way, I like FADEC, and it does do a great job most of the time (except at Heathrow with the dual "roll back" on final with the BA 777 and on dozens of airbuses over the years.) However, sometimes FADEC can't accelerate at it's full potential: if the airbus is in Level Change the FCP schedules a slow-spool up for pax comfort/tbo etc, and hits the trees, because:

It doesn't know the whole flying picture.

Only the pilot flying knows that. And only if he's not lulled to sleep riding around on autopilot watching the pfd FMA clicking through mindless modes that don't really keep the pilot informed as to what's happening unless you're staring right at it all the time. Which is why way back in 1995, AW&ST concluded that Airbus automation actually increases pilot workload, rather than reducing it as was advertised.

For me on the A-310's the answer was massive consumption of coffee making me a nervous wreck since I never knew when the damn thing was going to attempt a tail-slide or turn the wrong way, which it always did. It was like a bad student who had to be constantly watched.

Admiral and others are correct, 7 g's is GA stuff, my bad.

320 driver said:
No-one to my knowledge has ever 'fostered the habit' of flying on the back of the curve at low level (and I have been flying this machine for 16 years). It is a facility that can be useful in certain unusual situations. Flying the aircraft at SOP speeds (O, F, S) keeps you right in the flat part of the curve. If you are operating on the back of the curve then it is your poor practice, not a defect of the aircraft, its makers or its operating philosophy.
320 driver, you seem unaware of the early history of airbus. For example here was an accident by one of the chief pilots resulting in the first A320 fatalities:

Air France Flight 296 was a chartered flight of a newly-delivered fly-by-wire Airbus A320-111 operated by Air France. On June 26, 1988, as part of an air show it was scheduled to fly over Mulhouse-Habsheim Airport (ICAO code LFGB) at a low speed with landing gear down at an altitude of 100 feet, but instead slowly descended to 30 feet before crashing into the tops of trees beyond the runway. Three passengers died. The cause of the accident is disputed, as many irregularities were later revealed by the accident investigation. This was the first ever crash involving an Airbus A320

The official report states[1] the causes of the accident were:
  • Very low flyover height, lower than surrounding obstacles.
  • Very low speed, reduced to reach maximum possible angle of attack.
  • Engines idling during flight.
  • Late application of go-around power.
This combination led to the impact of the aircraft with the trees.
The Commission believed that if the descent below 100 feet was not deliberate, it may have resulted from failure to take proper account of the visual and aural information intended to give the height of the aircraft.

[edit] A320 operation anomalies

Third-party investigations into the crash dispute the official findings.[2] Captain Asseline asserted the altimeter read 100 feet (30 m) despite video evidence that the plane was as low as 30 feet (10 m). He also reported that the engines didn't respond to his throttle input as he attempted to increase power. The month prior to the accident, Airbus posted two Operational Engineering Bulletins indicating anomalous behaviour noted in the A320 aircraft. These bulletins were received by Air France but not sent out to pilots until after the accident:
[edit] OEB 19/1: Engine Acceleration Deficiency at Low Altitude

This OEB noted that the engines may not respond to throttle input at low altitude.
Air France Flight 296 - Wikipedia, the free encyclopedia

Also, you might want to read up on the recent splashdown of an A320 in Europe, and the dozens of alpha mode mishaps and accidents on the A300/A310's if you are unaware of them. Everybody and every property produces different flying experiences, I'm sure. But the cult of a "magic airplane", as another poster put it, used to be a big part of airbus culture and was dangerous imho.

Studies have shown, that humans are very poor at monitoring automation. Although man does not live by hand alone, hand flying proficiency is essential imho to avoid all these coupled disasters that plague the industry. If you hand-fly you actually reduce crew workload and increase safety imho. If you operate as well, with manual thrust, you are better prepared for unexpected events like AF447 when, at dark of night, outto drops out and throws it all in your lap . But how do you retain hand flying skills in an airplane which calculates your bank angle and back pressure for you, and just when you need it most falls into manual law?

CC
Captain-Crunch is offline  
Old 12th Dec 2010, 09:21
  #134 (permalink)  
 
Join Date: Aug 2007
Location: London, New York, Paris, Moscow.
Posts: 3,632
Likes: 0
Received 0 Likes on 0 Posts
read up on the recent splashdown of an A320 in Europe,
I'm sure the families of the deceased will appreciate your flippancy and inaccuracies.
glad rag is offline  
Old 12th Dec 2010, 11:35
  #135 (permalink)  
 
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes on 0 Posts
For me on the A-310's the answer was massive consumption of coffee making me a nervous wreck since I never knew when the damn thing was going to attempt a tail-slide or turn the wrong way, which it always did.
Sorry C.C., what are you talking about?

Thx
hetfield is offline  
Old 12th Dec 2010, 13:44
  #136 (permalink)  
 
Join Date: Aug 2007
Location: The sky
Posts: 337
Received 4 Likes on 1 Post
Captain Crunch,

How on earth do you attribute the BA777 dual rollback to the FADECS? The power loss was caused by fuel starvation due to ice in the fuel/oil heat exchangers. Your accuracy with this is as good as your Airbus knowledge.

FWIW (8000 hours, 50/50 Airbus / Boeing split), I much prefer the airbus GPWS pullup manouever. Give me max back stick compared to 20 degrees nose up any day. The Airbus climbs way way quicker. If you are ever in a situation where overloading the airframe in 'G' is the only thing that will save you then you should not be flying any aircraft, let alone a passenger one. If you ever want to see how fast an Airbus can pull up, watch this video.

YouTube - Airbus radical departure , Decolagem louca

I think there should be a forum rule, you can only comment on the Airbus once you have passed a type rating on one. Same for Boeings.

LD
Locked door is offline  
Old 12th Dec 2010, 15:15
  #137 (permalink)  
 
Join Date: Aug 2005
Location: fairly close to the colonial capitol
Age: 55
Posts: 1,693
Likes: 0
Received 0 Likes on 0 Posts
Mr. 320 Driver:

in the AA New York accident the rudder was being solely manipulated by the pilot. It was pilot error that generated the rudder input. The system had nothing to do with it.
On AA's A300's there is no data input from the rudder pedals. The DFDR records rudder movement only. Because the FO was the PF, it was assumed that the rudder was moving on his command.

Also, the DFDR traces were such a mess, the NTSB never published the raw traces and they had to cook the data in order to make any sense of it. This is partially due to the type of recorder the aircraft was equipped with which sampled rudder position at a rate of 2 points per second. Those units were later de-certified.

I am not summarily condemning the A300 in this accident, however I can say with confidence that the investigation was not among NTSB's finer moments.
vapilot2004 is offline  
Old 12th Dec 2010, 15:48
  #138 (permalink)  
 
Join Date: Sep 2008
Location: MI
Posts: 570
Likes: 0
Received 0 Likes on 0 Posts
I think there should be a forum rule, you can only comment on the Airbus once you have passed a type rating on one.
Ha...I can just imagine a type rating on a Bus. You probably have to learn how to type: ASDF ;LKJ, etc. Have they got CONTROL/ALT/DELETE on those things?

Last edited by DC-ATE; 12th Dec 2010 at 18:11.
DC-ATE is offline  
Old 12th Dec 2010, 15:52
  #139 (permalink)  
 
Join Date: Aug 2007
Location: The sky
Posts: 337
Received 4 Likes on 1 Post
Which is just the typical biased, unthinking and ill informed prejudice my suggestion would stop!

It's just another aircraft. A tool to do a job. Boeing use FBW now too, they were just slower to catch on.
Locked door is offline  
Old 12th Dec 2010, 15:55
  #140 (permalink)  
 
Join Date: Nov 2009
Location: Near Puget Sound
Age: 86
Posts: 88
Likes: 0
Received 0 Likes on 0 Posts
Could you miss the mountains in a pull up with this type of g-limiting and bank-limiting A320 FBW sidestick?
No, I don't think you can. FBW will save the airframe from overload first, but ultimately hit the rocks as I understand it.


I thought so too. Back in the late 80s, I got into a lengthy argument with Gordon Corps, an Airbus test pilot. I suggested a contest. We would each dive at the ridge, and when I said "Now" we would each pull, me at slightly under the 3.75 g ultimate and he at 2.5 g. Whichever one missed the ridge would win.

What I didn't realize (and Airbus didn't explain well) was that he could snap the stick back to the stop and get to 2.5 g's right now. I would have to ease the yoke back and would ultimately take a few seconds. Later ALPA promoted an evaluation and found that snapping the stick right back smartly actually had the airplane bottom out somewhat higher.

I've come to the conclusion that Airbus got that part right. In fairness to Boeing, while the limits in g's are "soft" they're hard enough that I think you'll get the same effect.

Let's all get over it. A well-designed envelope protection scheme really works. It's saved airplanes. In the twenty some years of commercial FBW, I don't know of any caused by g-envelope protection.

Goldfish
goldfish85 is offline  


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.