AMR 587 Airbus Crash (merged)
ex-Tanker
Join Date: Apr 2000
Location: Luton Beds UK
Posts: 907
Likes: 0
Received 0 Likes
on
0 Posts
I guess the first lesson for us all is that the rudder is a very powerful control. It can conceivably get you out of an upset situation but it can for sure get you into one too. It can also tear off the Fin (Vert Stab.) - as has been attested to by no less than John Tullamarine, the respected test pilot - if used in a cyclic manner. The RAF also issued warnings about using the flight controls in this manner on certain larger jet aircraft - particularly when being flown in formation.
Before going any further let us admit that we have little knowledge of the case in question and we weren't in the cockpit. What follows is surmise only on my part.
If the rudder is the last option in an upset, in my opinion it should be used. Vigorous use for "A little wake turbulence", however wouldn't be appropriate.
The comments from the Captain to the FO "Are you all right? " were during just such a cyclic useage period. This may have indicated a certain concern with the flight control inputs (as opposed to the atmospheric conditions.)
I have fairly often noticed a tendency for pilots to use rudder cyclically during approach in turbulent conditions (I am not talking about the alignment phase during the flare/touch down). This actually causes greater problems because an opposite aileron input is then required. The IAS difference from 140 odd knots to 260 knots and higher - as during climb out - causes a tremendous increase in the effect of control inputs.
The recent warning not to use the rudder in this manner still seems sensible.
Before going any further let us admit that we have little knowledge of the case in question and we weren't in the cockpit. What follows is surmise only on my part.
If the rudder is the last option in an upset, in my opinion it should be used. Vigorous use for "A little wake turbulence", however wouldn't be appropriate.
The comments from the Captain to the FO "Are you all right? " were during just such a cyclic useage period. This may have indicated a certain concern with the flight control inputs (as opposed to the atmospheric conditions.)
I have fairly often noticed a tendency for pilots to use rudder cyclically during approach in turbulent conditions (I am not talking about the alignment phase during the flare/touch down). This actually causes greater problems because an opposite aileron input is then required. The IAS difference from 140 odd knots to 260 knots and higher - as during climb out - causes a tremendous increase in the effect of control inputs.
The recent warning not to use the rudder in this manner still seems sensible.
Join Date: Mar 2000
Location: Arizona USA
Posts: 8,571
Likes: 0
Received 0 Likes
on
0 Posts
Hmmm, FewCloudy may just have hit the mark...many older guys who have been 'round for a long time (or...who were taught by same) know full well that large/rapid rudder inputs (machine or pilot induced) are NOT a good idea.
Look in the past guys for appropriate authority.
Seem to remember that Davies had it right in "Handling the Big Jets"... a primer for all younger jet airliner pilots.
Have loaned my copy to countless guys...good stuff enclosed.
Live and learn...
Look in the past guys for appropriate authority.
Seem to remember that Davies had it right in "Handling the Big Jets"... a primer for all younger jet airliner pilots.
Have loaned my copy to countless guys...good stuff enclosed.
Live and learn...
Guest
Posts: n/a
AMR 587 - The Real Issue!
> American denies Airbus ever formally warned that moving the rudder too
> much at high speed could cause it to come off. (Comment on AVweb)
The problem in this incident is that it did not happen at high speed. The whole flight lasted about two minutes from take off to crash. Obviously the aircraft was in post take off climb, probably following noise reduction procedure, and in that situation and short time could not possibly reach high speed!
I would expect ANY aircraft to be structurally capable of full control deflections at low speed, particularly at normal climb or approach speeds where wake turbulence and wind shear requiring strong control inputs are likely to occur. It is completely unacceptable that anyone should design, make, certify, operate and fly any aircraft incapable of this unless it is certified as Experimental and does not carry passengers, and then what about the risk to those on the ground. To do this with a public transport aircraft is sheer criminal negligence at the very least. I hope ALL of the culprits will be brought to account.
It is unacceptable to continue flying public transport aircraft in which the public's safety depends on a pilot in a moment of stress and urgency remembering not to tread too hard! This obvious structural weakness, which is openly admitted by the manufacturer, is no different from the Comet and Concorde crashes and for public safety demands the same response - the IMMEDIATE revocation of the Airworthiness Certificates of all transport aircraft with such a weakness. Will the various worldwide aviation authorities, particularly the FAA since the crash happened on their turf, have the moral fibre of the Brits or will passengers continue to be put at unacceptable and unnecessary risk because of political or other expediency?
While there is never a good time for addressing such a problem the worldwide aviation recession should lessen the impact on affected operators since there are many surplus aircraft they can lease to keep their operations functioning.
Pilots of affected aircraft should refer to the Transport Canada article and govern themselves accordingly: http://www.tc.gc.ca/civilaviation/sy...5/3-02/371.htm
> much at high speed could cause it to come off. (Comment on AVweb)
The problem in this incident is that it did not happen at high speed. The whole flight lasted about two minutes from take off to crash. Obviously the aircraft was in post take off climb, probably following noise reduction procedure, and in that situation and short time could not possibly reach high speed!
I would expect ANY aircraft to be structurally capable of full control deflections at low speed, particularly at normal climb or approach speeds where wake turbulence and wind shear requiring strong control inputs are likely to occur. It is completely unacceptable that anyone should design, make, certify, operate and fly any aircraft incapable of this unless it is certified as Experimental and does not carry passengers, and then what about the risk to those on the ground. To do this with a public transport aircraft is sheer criminal negligence at the very least. I hope ALL of the culprits will be brought to account.
It is unacceptable to continue flying public transport aircraft in which the public's safety depends on a pilot in a moment of stress and urgency remembering not to tread too hard! This obvious structural weakness, which is openly admitted by the manufacturer, is no different from the Comet and Concorde crashes and for public safety demands the same response - the IMMEDIATE revocation of the Airworthiness Certificates of all transport aircraft with such a weakness. Will the various worldwide aviation authorities, particularly the FAA since the crash happened on their turf, have the moral fibre of the Brits or will passengers continue to be put at unacceptable and unnecessary risk because of political or other expediency?
While there is never a good time for addressing such a problem the worldwide aviation recession should lessen the impact on affected operators since there are many surplus aircraft they can lease to keep their operations functioning.
Pilots of affected aircraft should refer to the Transport Canada article and govern themselves accordingly: http://www.tc.gc.ca/civilaviation/sy...5/3-02/371.htm
Join Date: Feb 2001
Location: A PC!
Posts: 594
Likes: 0
Received 0 Likes
on
0 Posts
But there was no grounding of the B737 for rudder hard-over (at least two fatal crashes) or the B747 for fuel pump explosions (TWA 800 was not caused by little green men, the US Navy or Airbus-employed saboteurs!).
It happens every time - the US based Boeing lovers start screaming "Airbus is unsafe" - just because Airbus are wiping the floor with Boeing.
All aeroplanes are potentially lethal - but fewer people died in this accident than as a result of p!ss-poor US DOMESTIC airline security last year. Maybe in bad taste, but realistic, folks.
Get real - get a life - get out more.
It happens every time - the US based Boeing lovers start screaming "Airbus is unsafe" - just because Airbus are wiping the floor with Boeing.
All aeroplanes are potentially lethal - but fewer people died in this accident than as a result of p!ss-poor US DOMESTIC airline security last year. Maybe in bad taste, but realistic, folks.
Get real - get a life - get out more.
Join Date: Jan 2002
Location: USA
Posts: 286
Likes: 0
Received 0 Likes
on
0 Posts
"The problem in this incident is that it did not happen at high speed. The whole flight lasted about two minutes from take off to crash. Obviously the aircraft was in post take off climb, probably following noise reduction procedure, and in that situation and short time could not possibly reach high speed! "
Gear and flaps were up and wake was encountered approx 0:1:27 after leaving the ground, alt 1840, speed 241K and accel. They were cleared to FL130.
Gear and flaps were up and wake was encountered approx 0:1:27 after leaving the ground, alt 1840, speed 241K and accel. They were cleared to FL130.
Join Date: May 2002
Location: Australia
Posts: 2,242
Likes: 0
Received 0 Likes
on
0 Posts
Just a question
Does the A340 have a similar rudder construction and design to the one on the AA Airbus300?
In the unlikely event of losing two on the same side with T/O or climb power set I can see the need for a fairly quick and large amount of rudder to correct yaw.
In the unlikely event of losing two on the same side with T/O or climb power set I can see the need for a fairly quick and large amount of rudder to correct yaw.
Regardless of how one feels about the design and certification standards, the fact remains that they are standards and represent a sizeable fleet. If some pilots feel that they can not adjust to these standards than they should simply stand down until such time that the standards are changed to accomodate them as the fleet retires due to old age.
The passengers deserve reassurance that it is safe to fly, not condemnation of the fleet that they fly on by people that were entrusted to understand and be trained to safely fly this fleet.
The passengers deserve reassurance that it is safe to fly, not condemnation of the fleet that they fly on by people that were entrusted to understand and be trained to safely fly this fleet.
Last edited by lomapaseo; 5th Nov 2002 at 01:07.
Union Goon
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes
on
0 Posts
During the investigation of 587 the entire AA fleet of all aicraft was looked at for wake encounters, CAT, and anything else that might have over stressed the tails.
DFDRs data from the incidents were analyzed. In the 800 aircraft fleet that comprises AA (not including the commuter jets) there were no incidents of design or Ultimate load being exceded in the Fokker, McDonnel Douglas or Boeing fleets. In the teeny tiny airbus fleet of 35 aircraft there were 7 excedences of design loading of the tail (including two cases of autopilot malfuntions that caused the rudder pedals to continually move back and forth. It was impossible to disengage the rudder channel from the autopilot) and two excedences of ultimate load (587 and the other tail that is still removed in Tulsa, which turned out to be cracked even though several inspections had not found the cracks untill the tail was removed from the aircraft and ultrasounded)
Furthermore throughout the fleet more examples were found at FEDEX, a German airline, and on a swissair 330. (and yes the tails are very similar)
This is a serious problem that needs to be fixed. This is not about protectionism. Airbus has done absolutely ZERO to address this known problem. Atleast boeing comes up with a workaround untill a permanent fix can be implemented.
Cheers
Wino
DFDRs data from the incidents were analyzed. In the 800 aircraft fleet that comprises AA (not including the commuter jets) there were no incidents of design or Ultimate load being exceded in the Fokker, McDonnel Douglas or Boeing fleets. In the teeny tiny airbus fleet of 35 aircraft there were 7 excedences of design loading of the tail (including two cases of autopilot malfuntions that caused the rudder pedals to continually move back and forth. It was impossible to disengage the rudder channel from the autopilot) and two excedences of ultimate load (587 and the other tail that is still removed in Tulsa, which turned out to be cracked even though several inspections had not found the cracks untill the tail was removed from the aircraft and ultrasounded)
Furthermore throughout the fleet more examples were found at FEDEX, a German airline, and on a swissair 330. (and yes the tails are very similar)
This is a serious problem that needs to be fixed. This is not about protectionism. Airbus has done absolutely ZERO to address this known problem. Atleast boeing comes up with a workaround untill a permanent fix can be implemented.
Cheers
Wino
Last edited by Wino; 7th Nov 2002 at 20:00.
Join Date: Aug 1998
Location: The World
Posts: 388
Likes: 0
Received 0 Likes
on
0 Posts
I received this Avweb Avflash yesterday, rather interesting...
"587 HEARING POLARIZES: The public hearing into the crash of an American
Airlines Airbus last November has degenerated into a polarized blamefest
over rudder deflection. There's now no argument that five full back-
and-forth deflections of the A300-600's rudder caused the tail to rip
off American Airline's Flight 587, precipitating its crash in New York
last November 12, which killed all 260 aboard and five on the ground.
Airbus maintains it told American Airlines that its training on use of
the rudder was faulty and indicated that communication implies the
carrier knew such a failure was possible. American denies Airbus ever
formally warned that moving the rudder too much at high speed could
cause it to come off."
"587 HEARING POLARIZES: The public hearing into the crash of an American
Airlines Airbus last November has degenerated into a polarized blamefest
over rudder deflection. There's now no argument that five full back-
and-forth deflections of the A300-600's rudder caused the tail to rip
off American Airline's Flight 587, precipitating its crash in New York
last November 12, which killed all 260 aboard and five on the ground.
Airbus maintains it told American Airlines that its training on use of
the rudder was faulty and indicated that communication implies the
carrier knew such a failure was possible. American denies Airbus ever
formally warned that moving the rudder too much at high speed could
cause it to come off."
Union Goon
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes
on
0 Posts
No one has yet proved that the pilots moved the rudder at all!
It has been proved that the rudder load limiter did not limit the rudder to design or ultimate load however. Further it has been proven that 2 other incidents have occured where the autopilot did not declutch from the rudder channel exceded design loading on the tail. In both of those cases the rudder pedals moved on their own for the entire duration of the event.
Most certainly the statement from the Airbus AOM that that rudder load limiter works by physically limiting the travel of the rudder so as not to over stress or over control the aircraft at high speeds has been proven false.
The Airbus alternate gear extension checklist calls for alternating rudder inputs to induce back and forth sideslips to lock the gear in place! Obviously according to airbus testimony we can't do that either. I guess I should just belly it in if I have a hydralic failure.
Cheers
Wino
It has been proved that the rudder load limiter did not limit the rudder to design or ultimate load however. Further it has been proven that 2 other incidents have occured where the autopilot did not declutch from the rudder channel exceded design loading on the tail. In both of those cases the rudder pedals moved on their own for the entire duration of the event.
Most certainly the statement from the Airbus AOM that that rudder load limiter works by physically limiting the travel of the rudder so as not to over stress or over control the aircraft at high speeds has been proven false.
The Airbus alternate gear extension checklist calls for alternating rudder inputs to induce back and forth sideslips to lock the gear in place! Obviously according to airbus testimony we can't do that either. I guess I should just belly it in if I have a hydralic failure.
Cheers
Wino
Trash du Blanc
Join Date: Mar 2001
Location: KBHM
Posts: 1,185
Likes: 0
Received 0 Likes
on
0 Posts
Some enlightenment, please:
the previously posted Airbus graph shows "Pedal Command + YD."
Is that just pedal position? Do the pedals shadow the YD inputs, hence the "+YD" label?
Or is the pilot the only source of pedal movement?
the previously posted Airbus graph shows "Pedal Command + YD."
Is that just pedal position? Do the pedals shadow the YD inputs, hence the "+YD" label?
Or is the pilot the only source of pedal movement?
Psychophysiological entity
During multiple rudder deflections, for whatever reason, what is known about the differential air pressures within the fixed vertical structure. The lack of adequate venting that has been a problem in the past, might exacerbate the stresses during side loadings, especially if it changes rapidly several times in phase with the mechanical loads
Union Goon
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes
on
0 Posts
If the autopilot commands a rudder movement the pedals will move as well. Normally this only happens with the slats extended.
However two previous failures of the clutching mechanism left the autopilot incontroll and moving the rudder and pedals even though the autopilot was completely disconnected (disconnect buttons pushed, paddles down, autopilot still moving the rudder pedals, aircraft wildy yawwing back and forth)
Cheers
Wino
However two previous failures of the clutching mechanism left the autopilot incontroll and moving the rudder and pedals even though the autopilot was completely disconnected (disconnect buttons pushed, paddles down, autopilot still moving the rudder pedals, aircraft wildy yawwing back and forth)
Cheers
Wino
Join Date: Aug 2001
Location: Asia
Posts: 381
Likes: 0
Received 0 Likes
on
0 Posts
Wino
You're clear description of the A300-605 sytem is frightening to say the least.
It would logically follow that anyone with 1,000 hours on the type would avoid even tickling the rudder pedals if at all possible.
To suggest that there were multiple inputs of opposite rudder is to imply a suicide conspiracy which is clearly nonsensical.
What has been done by type operators and Airbus in the meantime to address this issue?
You're clear description of the A300-605 sytem is frightening to say the least.
It would logically follow that anyone with 1,000 hours on the type would avoid even tickling the rudder pedals if at all possible.
To suggest that there were multiple inputs of opposite rudder is to imply a suicide conspiracy which is clearly nonsensical.
What has been done by type operators and Airbus in the meantime to address this issue?
Join Date: May 2002
Location: Canada
Posts: 155
Likes: 0
Received 0 Likes
on
0 Posts
TO: moggie
"But there was no grounding of the B737 for rudder hard-over (at least two fatal crashes)."
True, but despite the rudder hard-overs none caused the fin to fail and that is the important issue here that should and must be resolved. Also the 737 problem was an extremely difficult mystery to solve and it was only through many years/decades of effort by ALPA that the truth was discovered and the blame shifted from the pilots. There appears to be no mystery to the A300 problem and safety must be above jingoism in this industry.
"But there was no grounding of the B737 for rudder hard-over (at least two fatal crashes)."
True, but despite the rudder hard-overs none caused the fin to fail and that is the important issue here that should and must be resolved. Also the 737 problem was an extremely difficult mystery to solve and it was only through many years/decades of effort by ALPA that the truth was discovered and the blame shifted from the pilots. There appears to be no mystery to the A300 problem and safety must be above jingoism in this industry.
Join Date: Jan 2002
Location: USA
Posts: 286
Likes: 0
Received 0 Likes
on
0 Posts
Wino
Agree, according to AirBus, they stated that the correct procedure for extending the gear was the alternating rudder inputs you refer. I would assume gear extension speed is something short of 250K, but could certainly be high enough to cause potential structural damage if excessive rudder is applied. Accordingly, what speed restriction is noted for this alternating rudder procedure. During the same exchange, AA indicated their procedure was to bank the airplane rather than yaw.
Agree, according to AirBus, they stated that the correct procedure for extending the gear was the alternating rudder inputs you refer. I would assume gear extension speed is something short of 250K, but could certainly be high enough to cause potential structural damage if excessive rudder is applied. Accordingly, what speed restriction is noted for this alternating rudder procedure. During the same exchange, AA indicated their procedure was to bank the airplane rather than yaw.
Union Goon
Join Date: Feb 2000
Location: New Jersey, USA
Posts: 1,097
Likes: 0
Received 0 Likes
on
0 Posts
Yes we modified the procedure to increase g loading via coordinated turns (bank the aircraft) to try and extend the gear and get the overcenter locks that way. It is not a good solution however, just the best we can do.
If there was some sort of major problem that forced us to be clean at the same time (can't be total loss of hydraulics because would make the aircraft unfliable, but you never know multiple failures DO happen) green dot would be around 235 knots or so at max landing weight, so we would be doing the gear extension procedure up around that speed.
Furthermore, as it is an unusual procedure that requires moving your seat and sticking a handle in the ground and turning it some recommend doing it in the holding pattern or before the approach. Again, 235 knots or so is a good clean maneuvering speed with a full load of pax and not much fuel.... If it was a return after takeoff (no dump for upto 120,000 lbs of fuel) green dot could be well above 250.
Airbus posts no speed limit other than the normal gear door limits (270 knots) and says nothing about slowing down to apply rudder. The manual says the rudder load limiter will protect us in its description of the rudder load limiter.
Cheers
Wino
If there was some sort of major problem that forced us to be clean at the same time (can't be total loss of hydraulics because would make the aircraft unfliable, but you never know multiple failures DO happen) green dot would be around 235 knots or so at max landing weight, so we would be doing the gear extension procedure up around that speed.
Furthermore, as it is an unusual procedure that requires moving your seat and sticking a handle in the ground and turning it some recommend doing it in the holding pattern or before the approach. Again, 235 knots or so is a good clean maneuvering speed with a full load of pax and not much fuel.... If it was a return after takeoff (no dump for upto 120,000 lbs of fuel) green dot could be well above 250.
Airbus posts no speed limit other than the normal gear door limits (270 knots) and says nothing about slowing down to apply rudder. The manual says the rudder load limiter will protect us in its description of the rudder load limiter.
Cheers
Wino
Join Date: Jun 2002
Location: Houston, Texas
Posts: 25
Likes: 0
Received 0 Likes
on
0 Posts
Very useful information here. Safety is well served by honest public discussion and debate. I've several thoughts here:
Airbus certainly has no monopoly on less than stellar behaviour during post accident certification issues. One only needs to remember how Boeing responded during the Palm 90 investigation, particularly in regards to pre-loading and flying various mathematical models for leading edge contamination on the sim prior to turning it over to the NTSB. (Nance 1986)
I'm not much on aerodynamics, but does'nt AOA have a direct relationship on which control surface provides the most roll authority? At higher AOA we reach a speed where the aileron/spoiler create less of a rolling moment than the rudder. (Crossover speed) As has been pointed out, we do not yet even know if the crew used the rudder. I mention crossover speed to the never-use-the-rudder crowd. I know some of you folks have a lot more flying time than I do, but do you really think you're going to recover an aircaft at >90 degrees, loaded up without carefull but firm use of both forward stick and full rudder followed by coordinated aileron spoiler?
United had an A319 in December of 1999 that had a number one uncontained failure in a 800' 25 degree left turn. The AC rolled to 45 degrees. the Captain (ex-fast jet type) unloaded to 1/2G and STOOD on the right rudder. I'm not familiar with what envelope protection the A319 would have provided him rudder-wise, but he recovered with 200 AGL. I bring this incident up because I think we as an industry are not yet in agreement on what constitutes appropriate use of rudder during an upset. Based on incidents such as this one, how can you say that aileron/spoiler only would have been appropriate? I simply don't see how he could have recovered without aggresive use of the rudder. I don't bring this up to insult anyone, I'm here to become a better educated aviator. Can anyone in the never-use-the-rudder crowd please explain to me how their philosophy relates to crossover speed? At high G/angle of bank situations we have little roll capability. I'm going to do whatever gets the lift vector point up quickly. RUDDER.
On a less related note concerning the 737 un-commanded yaw controversey: has anyone seen the Boeing 737 footage of the wing tip vortices encounters they flight-tested behind the FAA's 727? With autopilot disengaed and no hands or feet on the controls, there was a +70 degree upset. Both the Pittsburgh and Colorado Springs accidents recorded roll rates well in excess of what deliberate, full deflection aileron and rudder inputs would have resulted in. Food for thought.
There are aviators far more talented than I will ever be, who for one reason or another got to meet Elvis. There but for the grace of god go we.
Cheers,
DAN
Airbus certainly has no monopoly on less than stellar behaviour during post accident certification issues. One only needs to remember how Boeing responded during the Palm 90 investigation, particularly in regards to pre-loading and flying various mathematical models for leading edge contamination on the sim prior to turning it over to the NTSB. (Nance 1986)
I'm not much on aerodynamics, but does'nt AOA have a direct relationship on which control surface provides the most roll authority? At higher AOA we reach a speed where the aileron/spoiler create less of a rolling moment than the rudder. (Crossover speed) As has been pointed out, we do not yet even know if the crew used the rudder. I mention crossover speed to the never-use-the-rudder crowd. I know some of you folks have a lot more flying time than I do, but do you really think you're going to recover an aircaft at >90 degrees, loaded up without carefull but firm use of both forward stick and full rudder followed by coordinated aileron spoiler?
United had an A319 in December of 1999 that had a number one uncontained failure in a 800' 25 degree left turn. The AC rolled to 45 degrees. the Captain (ex-fast jet type) unloaded to 1/2G and STOOD on the right rudder. I'm not familiar with what envelope protection the A319 would have provided him rudder-wise, but he recovered with 200 AGL. I bring this incident up because I think we as an industry are not yet in agreement on what constitutes appropriate use of rudder during an upset. Based on incidents such as this one, how can you say that aileron/spoiler only would have been appropriate? I simply don't see how he could have recovered without aggresive use of the rudder. I don't bring this up to insult anyone, I'm here to become a better educated aviator. Can anyone in the never-use-the-rudder crowd please explain to me how their philosophy relates to crossover speed? At high G/angle of bank situations we have little roll capability. I'm going to do whatever gets the lift vector point up quickly. RUDDER.
On a less related note concerning the 737 un-commanded yaw controversey: has anyone seen the Boeing 737 footage of the wing tip vortices encounters they flight-tested behind the FAA's 727? With autopilot disengaed and no hands or feet on the controls, there was a +70 degree upset. Both the Pittsburgh and Colorado Springs accidents recorded roll rates well in excess of what deliberate, full deflection aileron and rudder inputs would have resulted in. Food for thought.
There are aviators far more talented than I will ever be, who for one reason or another got to meet Elvis. There but for the grace of god go we.
Cheers,
DAN