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

AA 587 Final Report

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

AA 587 Final Report

Old 6th Dec 2004, 21:48
  #1 (permalink)  
Thread Starter
 
Join Date: Jul 2000
Location: U.S.A.
Posts: 474
Likes: 0
Received 0 Likes on 0 Posts
AA 587 Final Report

The NTSB has issued the final report for AA 587 and is available at:

http://www.ntsb.gov/publictn/2004/AAR0404.htm
Shore Guy is offline  
Old 6th Dec 2004, 22:39
  #2 (permalink)  
 
Join Date: Mar 2000
Location: Arizona USA
Posts: 8,571
Likes: 0
Received 0 Likes on 0 Posts
Page 159, section 3.1, findings

Are indeed quite damning of American Airlines flight training procedures, the particular handling pilots' (the First Officer) actions, and indeed calls into question the whole idea of jet upset training, as followed by American Airlies at that time.

Lessons so long ago learned in the very early days of large jet transport flying are, or have been, completely ignored by American Airlines.

Can we really be surprised at all this?
I certainly am not.
411A is offline  
Old 7th Dec 2004, 02:05
  #3 (permalink)  
 
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes on 0 Posts
In section 1.6.2.3 "A300-600 Rudder Control System Design Compared to Other Airplanes", on pages 27 thru 29, Tables 4 and 5.

It appears that the rudder control system and limiter of the A320, A330 and A340 are almost as sensitive at higher speeds as the A300-600. I find this shocking and almost unbelievable. It seems possible that this type of accident could happen almost as easily on these aircraft, as on the A300-600.

Please tell me I'm reading this information incorrectly, or the FBW system can protect the pilots.
Flight Safety is offline  
Old 7th Dec 2004, 23:32
  #4 (permalink)  
 
Join Date: Nov 2004
Location: Here, there, and everywhere
Posts: 1,117
Likes: 0
Received 12 Likes on 7 Posts
"Lessons so long ago learned in the very early days of large jet transport flying are, or have been, completely ignored by American Airlines."

Would be interested in hearing what some of these lessons are. Let's spread the knowledge. I have no doubt you have seen plenty over the years. Its always interesting to hear some stories from the uh... old timers.
punkalouver is offline  
Old 8th Dec 2004, 02:14
  #5 (permalink)  
 
Join Date: Apr 1999
Location: Canada
Age: 82
Posts: 91
Likes: 0
Received 0 Likes on 0 Posts
Large Jet, Small Jet, Any Jet, for that matter Any Airplane:

Basic training has always taught that you deflect the flight controls with great discretion especially in the higher speed range of the envelope. Doesn't matter whether it is a Cessna or an SST; full deflection will stress/can over-stress the airframe. Sure, in a full stall in a C-172, you will use full rudder to counteract yaw and avert a spin but in a large jet ?? At speed ???

Yes, most modern transports have "built-in" protections from such inputs, either mechanical stops or hydraulic limiters, but it is really a stretch to see a major carrier (AA) teaching pilots to yard the rudder from stop to stop (limiters notwithstanding) just to cope with turbulence. And also .... are we talking turbulence or "jet upset" - there is a big difference. My experience with jet transports ranging from the DC-9 to the B-747 is that one is best advised to just hang on and damp out the motion (if hand flying) or let the autopilot cope if in auto-flight. Intervening with aggressive inputs, especially on the rudder, is counter to everything that I was taught and that experience enforced.


While most of the early jet transports needed all the rudder for an engine failure at the critical speed, most of the more modern ones are "over-ruddered" and only require a portion of that available - you can actually fly through an engine failure at the critical speed using aileron alone. It's not pretty, but it works. So why would an airline teach stomping on the rudder to cope with turbulence? Especially at speeds well above the stall.

I do not always agree with 411A, especially when he knocks AA, but he does have a point here. They/he broke that airplane.
Idle Thrust is offline  
Old 8th Dec 2004, 04:28
  #6 (permalink)  
Union Goon
 
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes on 0 Posts
No one was taught to STOMP on the rudder.

The A300 has had SEVERAL Pilot induced ocilations in the Yaw regime by several different airlines.

That makes the AIRCRAFT unique throughout the airline industry. NO OTHER type of aircraft has a similar history. And the incidents are NOT confined to AA service.

CHeers
Wino
Wino is offline  
Old 8th Dec 2004, 05:49
  #7 (permalink)  
 
Join Date: Jul 2000
Location: Down south, USA.
Posts: 1,594
Received 9 Likes on 1 Post
Snoop

Wino: Did the pilots at (formerly East German) Interflug, Air France and other companies which experienced such rudder oscillations ALSO stomp on the rudders? Can anyone find access to the info, other than Airbus, Inc? Can any Air France or Lufthansa, Condor or Eurowings (G-Wings etc) pilots out there with contacts from the old "DDR" find out whether such data is available? Or would this not be desireable, based on pride in local/shared technology?

If they allegedly stomped more than once on their rudders, then this is quite interesting. If those pilots, as a group, did NOT push aggresively on both rudder pedals, then this would be more than just interesting. Maybe the authorities will keep this "Pandora's box" sealed for many years-or shred the data.
Ignition Override is offline  
Old 8th Dec 2004, 08:12
  #8 (permalink)  
 
Join Date: Sep 2001
Location: 38N
Posts: 356
Likes: 0
Received 0 Likes on 0 Posts
One has a sense the AA587 report conclusion is driven more by the political goal of detente than by a true reading of de facts.
arcniz is offline  
Old 8th Dec 2004, 08:26
  #9 (permalink)  
Just a numbered other
 
Join Date: Feb 2000
Location: Earth
Age: 72
Posts: 1,169
Received 0 Likes on 0 Posts
fish

That's right, arcniz.

They find that if (for whatever reason of incompetence or poor training) the pilot reverses the rudder to its full stops several times in 6.5 seconds, it will fail despite withstanding more than its design and certification requirements.

Very political
Arkroyal is offline  
Old 8th Dec 2004, 12:31
  #10 (permalink)  
Union Goon
 
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes on 0 Posts
This particular model of airbus has a trap laid out in it and no one wants to accept it. Ark Royal. Do you really believe the pilot INTENDED to reverse the rudder quickly several times in few seconds? I guarantee you he did not intend to do that. If he didn't INTEND to do it, then its a problem with the airplane. The controlls of an airplane should be intuitive. They should respond in a predictable manner. In this area of the flight envelope of the A300 they most certainly do not.

while it is exceptionally rare that you should need to use the rudder at faster than landing speeds, it DOES happen. Once in a career. Possibly more rarely than an engine failure.

The A300 is the ONLY airplane in the world where it is not POSSIBLE to apply cooridinated rudder should you need to for what ever reason.,

If you TRY and use coordinated rudder in the A300 at 250 knots it is a virtual guarantee that you will start banging the stops.

If it is your opinion ARK royal that the rudder should NEVER be used at that speed, then it should be locked out.

I have several thousand hours in the aircraft, at the airline and took all the courses that Sten Molin and Ed States were alledged to have taken. I guarantee you they are NOT what you THINK you know from TV or news, and that Airbus did a masterfull job of covering up a REAL defficiency in their aircraft.

ARKROYAL and ALL the others here, if you can show me ONE just one aircraft out of the 10s of thousands of transport category jets worldwide that excede its rudder load limits in a similar manner to 587 I will accept your premise of pilot error.

On the other hand out of 300 or so A300600rs world wide, I can show you 4 at 3 different airline that have had similar events.


Thats a PATTERN. Well documented and VERY serious a real handling problem with a small fleet of aircraft.

I can lay out HOW the aircraft has this problem. The break out force (Force required to get the rudder moving at all) is MUCH TO HIGH RELATIVE TO THE FORCE TO REACH THE STOP.

The Rudder sensitivity increases by an order of magnitude with speed.

Airbus arrogence is such that they would rather kill more people than admit they made a mistake and implement a cheap and effective fix. All the AA bashers are either simply yank bashing or taking the easy way out. Certainly not the first time we have seen that in aviation.

AA operates over 800 Jets of MD80 size or larger. If it was an AA handling issue, then surely there would have been events in 767s, 777 MD80s, MD-11 DC-10s etc... Yet there is none. Explain that one? Yet the SMALLEST fleet at AA (35 A300 jets) had two events. the other 750+ had none. HMMMMM statitstics don't bear out any of the Yank/AA bashers.

When someone shows me problems with other AA aircraft handling, or a POI of ANY type of transport category of a different aircraft type about the Yaw Axis, ANYWHERE in the world I will recant.

Cheers
Wino

Last edited by Wino; 8th Dec 2004 at 12:42.
Wino is offline  
Old 8th Dec 2004, 13:40
  #11 (permalink)  
I've only made a few posts so I don't feel the need to order a Personal Title and help support PPRuNe
 
Join Date: Sep 2000
Posts: 207
Likes: 0
Received 0 Likes on 0 Posts
Danger

Wino, whilst you put a good case out for the technical problem with the rudder could you please do us all a favour and stop it with your paranoia about "yank bashing". The biggest "yank basher" on here is none other than your own 411a... a "yank" no less!
cargo boy is offline  
Old 8th Dec 2004, 13:58
  #12 (permalink)  
 
Join Date: Sep 2001
Location: Toronto
Posts: 2,553
Received 24 Likes on 13 Posts
I can lay out HOW the aircraft has this problem. The break out force (Force required to get the rudder moving at all) is MUCH TOO HIGH RELATIVE TO THE FORCE TO REACH THE STOP.
In Airbus the ratio is 1.45 -- in Boeings and MD, the ratio ranges from 3 to 6.5.

AA's part in this was rigging the simulator with a different rudder response model than was in the a/c.

The report makes it clear that the FO was likely surprised by the a/c response to his rudder input as the pedals at that speed give you either nothing or full on.

The two other incidents where the FO was aggressive with the rudder were in 727s which have rudder ratio changers and a maximum deflection to breakout force TWICE that of the A300-600.

That training and experience were then brought to an a/c where:
Tests were also conducted in which the subjects were instructed to move the control wheel and rudder pedal to 50 percent of their available range. The tests showed that the pedal force applied during the 50-percent condition resulted in full rudder travel, even though that force was one-half of the force applied at the 100-percent condition.
Page 75 (89 of 212) of the report.

For those still unable to grasp the human factor significance of a low maximum deflection to breakout force ratio, lets just call it rudder snatch.
RatherBeFlying is offline  
Old 8th Dec 2004, 16:08
  #13 (permalink)  
Union Goon
 
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes on 0 Posts
Cargo boy,

Change yank to AA bashing and you have 411a to a T. Eitherway I am describe prejudice that gets in the way of looking at the facts of the case.

In this case 411a happens to be one of the biggest bigots on the board. It is just difficult to categorize it so he falls under the yank bashing category. Lets call him a self hating american.

Cheers
Wino
Wino is offline  
Old 8th Dec 2004, 18:46
  #14 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 9 Likes on 5 Posts
Before Wino gets sent the full FAA / JAA list of all commercial aircraft, ‘that if flown inappropriately could exceed the rudder load limits’, just reflect that aircraft are not designed or certificated to be foolproof. The rationale is that fools do not fly aircraft, but it also presupposes that crews have some understanding of the aircraft design and certification that enable a safe operation; training provides this knowledge.

There are many assumptions made about pilot performance and behaviour in certification, some unfortunately have found to be wanting, as in this instance. There are still many other assumptions that crew need to know in order to prevent them breaking an aircraft; i.e. we assume that pilots will flare to land the aircraft, if they don’t most aircraft will be structurally weakened. That’s obvious, but who until recently remembered that Va was the limit speed for full control deflection … and what is Va.

The issue is not necessarily the rudder sensitivity, but that the rudder was cycled quickly between maximum values, causing dynamic over-loading of the fin. A single, quick maximum deflection (within Va) will not cause any harm to the structure. If crew’s were to use other aircraft controls in a cyclic manner, stop to stop with reversals, and it can be done in some conventional aircraft, then structural overload and failure is quite possible.

If you do something stupid in an aircraft it may break, the essential knowledge for pilots is to know all of the stupid things and avoid them.

“Somebody does somethin' stupid, that's human. They don't stop when they see it's wrong, that's a fool.” - Elvis Presley.

Because of this accident, it is probable that the certification requirements will be amended, either to eliminate the certification oversight or to provide tolerance in the event of crew abuse, but will this action result in the aircraft being closer to foolproof or is it just accepting that more fools are flying?
--------------------
Airspeed and Upwardness
safetypee is offline  
Old 8th Dec 2004, 20:48
  #15 (permalink)  
 
Join Date: Mar 2000
Location: Arizona USA
Posts: 8,571
Likes: 0
Received 0 Likes on 0 Posts
Let us presume for the moment that the A300-605R that American has in their fleet, is as bad as Wino suggests.

OK, now considering that American Airlines ordered 35 of 'em, and American Airlines carefully evaluated the design, and its requirements for use in their fleet, and had their training Captains sent to TLS to attend the manufacturers course, so that these same American Airlines training Captains could come back and report on any problems encountered, and pass this information on to American Airlines line crews, so that no untoward difficulties would occur...well you get the point, I'm sure.

The question remains, why did not American Airlines find out about any special problems that the particular type had, and if not, why not?

You will note many references to American Airlines in the above, and it is there for a reason...American Airlines sets a very poor standard for itself, which it consistantly fails to achieve, at least in the flight deck crew training department.

IE: not many bright bulbs in the lamp in Dallas.

Good grief, and these folks call themselves a 'large scheduled airline'...phooey.
411A is offline  
Old 8th Dec 2004, 21:16
  #16 (permalink)  
Union Goon
 
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes on 0 Posts
411a,

Much of the information REQUIRED to do what you suggest was NOT provided by Airbus. Sorry dude, but them's the facts...

The other cases of overloading the tail did NOT come to light untill the NTSB got them.

AA's manuals were in accordance with the factory manuals. I have a copy of the factory manuals on CDr, and they incorrectly describe the rudder load limiter system as would be applied to this case.

Furthermore Airbus's own published procedures had a case calling for "alternating sideslips" in the event of a problem during manual extension. To the best of my knowledge all manual gear extensions were so far trouble free, but sooner or later this would have come up there as well.

That procedure as since been DRAMATICALLY modified.

This is a problem that was somehow missed by the engineers. But rather than try and correct it, (which can be done with a software change) Airbus is gonna save face till the end and claim there is nothing wrong with the airplane, inspite of the overwhelming evidence.

411a, If AA is as bad as you say, how come this problem hasn't reared its head in any of the other aircraft in its fleet (which number more than 20 times as many)

Cheers
Wino
Wino is offline  
Old 8th Dec 2004, 21:42
  #17 (permalink)  
swh

Eidolon
 
Join Date: May 2001
Location: Some hole
Posts: 2,170
Received 24 Likes on 13 Posts
Dont let fact get in the way of a good story

Fact : American Airlines has completed in excess of 17 million flights, its one of the worlds safest airlines. Whilst a number of airlines in North America are statistically safer than American Airlines, on a global scale, only a few other large established airlines are statistically safer than American, i.e, British Airways, Lufthansa, All Nippon, Mexicana Airlines, and Qantas, however if you add the number of flights by all these airlines together they will only just get over the number of flights that American have completed.

Fact : No one here can attest to what the pilot intended to do, or what training he received.

Fact : The structure failed after loads were induced in the structure by pilot control inputs.

Fact : The loads induced to cause structural failure exceeded FAA certification requirements.

Fact : The A300 has had less incidents per number of flights than the B747, and has killed fewer people.

swh is offline  
Old 8th Dec 2004, 22:06
  #18 (permalink)  
Just a numbered other
 
Join Date: Feb 2000
Location: Earth
Age: 72
Posts: 1,169
Received 0 Likes on 0 Posts
fish

Wino,

I refer you to safetypee's excellent post.

The guy had a history of misuse of the rudder. AA's training included use of rudder where it was not needed, the sim excercise was totally unrealistic, and instilled a fear of wake encounters which was misplaced and excessive.

This aeroplane crashed because it was mishandled
Arkroyal is offline  
Old 9th Dec 2004, 14:00
  #19 (permalink)  
 
Join Date: Jul 2001
Location: Germany
Posts: 177
Likes: 0
Received 0 Likes on 0 Posts
Angel

SAFETYPEE and SWH
Well said. Could'nt phrase it better.

411a
neglecting your special love-relation concerning American Airlines all your other remarks how to handle or mishandle a large jet are appropriate IMHO.

WINO:
As you know there are two europeen airlines (AF and LH) which operated a number of A 300/306/R/310-200/300 in the past -partially until now- in huge numbers. Compared to that a fleet of 35 AC with AA is a "small" fleet.
Strolling through the floors/training departements/simulator briefing rooms and and talking to old buddies and presenting
some of your statements in different threads at PPRuNe most
people are NOT AMUSED.

As insiders know both incidents (A310) at MOW and Paris
had nothing to do with "oversensivity" or "breakout force" of the "unique" rudder system but everything with gross mishandling after pulling the TOGA lever unintentionally and missing knowledge of the consequences by not disconnecting the autopilot. This brave working horse A 310
at MOW even recovered itself 3 times flying on a vertical S flight path between full stall and x degree nose down pretty close to the ground. Stomping the rudder during this unusual handling overstressed the vertical stabilizer which was found out much later.
In this case the crew finally disconnected the AP, recovered and landed uneventful.

Your repeated statement of so many other airlines suffering problems with the A 300 rudder system or "pilot induced occilations in the yaw damper regime" are not factual.

BTW no repeat no european airline train or trained their crews the way AA did: Upset training scenarios in simulators outside their envelop refering to wake turbulance was no option on this side of the pond.

Regards
Captain104 is offline  
Old 9th Dec 2004, 14:53
  #20 (permalink)  
Dr Dave
Guest
 
Posts: n/a
Can I make an analogy to illustrate my view of this situation:

If I am driving my 4x4 at 70 mph (quite an achievement actually in this case...), and choose to turn the steering wheel quickly through two complete turns, then the vehicle will probably roll over and crash. If I do the same thing in my Mercedes 'C' class it almost certainly won't.

Now, I have been on an official training course run by the 4x4 manufacturer. At no time on this course was I told not to do this. However, it is up to me to know that I should not turn the wheel rapidly at high speed in this vehicle. If I do, it is not the manufacturers fault - they design a vehicle to do a certain job within a certain envelope. If I choose to take the vehicle outside that envelope then that is, frankly, my fault and the consequences are my problem, not theirs.

It is not an excuse to say that my Merc would not roll. They are different vehicles with different designers. As a driver it is up to me to know what the limitations of each vehicle are. Of course the 4x4 manufacturer could put in some fancy electronics to stop you doing this (perhaps a device that controls the powered steering according to speed). But, there is no need so long as drivers are sensible.

Of course, if the manufacturer told me that it was definitely safe to turn the steering wheel through two complete revolutions when it is not, then they would have to carry some (but not all) of the responsibility.

It seems to me that in the case of the Airbus A300, we have an aircraft that has been around for 30 years. Only one pilot has ever managed to snap the rudder off, and he did that by taking the aircraft significantly outside the envelope. It appears that he did that in part because his training led him to believe that this was an appropriate thing to do in those circumstances.

In this context, the fault just does not lie with the manufacturer.

So AA did not have this problem with their other fleets. Frankly, so what? The training for any fleet should be tailored to the needs of that fleet. Perhaps for the other aircraft types it was. For the A300 it appears to me at least that it probably wasn't.

Dr Dave
 

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.