PDA

View Full Version : AA Flight 587 accident at JFK:


wsherif1
9th Jun 2003, 16:32
All the Ph.D's and the Aero. Engineers at the AA 587 NTSB hearing, have not figured out the probable cause, even after seeing the picture of the vertical stabilizer showing the three rudder actuators completely severed from the rudder!

Airbus states flatly that the rudder was pilot commanded.! The rudder movements were so rapid, and in opposite directions, that no pilot could or would operate the rudder in such a manner.!

At some point in time the 0.3, 0.4 and the 0.8 G forces from the rotating vortices, of the B747, struck the vertical fin and rudder, BROADSIDE, and severed the rudder linkage to the actuators.

The rudder is now freely floating and moves in the direction of the wind shear forces. These wind shear forces are also striking the vertical fin, BROADSIDE, and now we have an additional, and very effective, flight control surface.

The large, two engine jet transport design requirement for a large vertical stabilizer, for engine out on take-off control, has created a very large, weather-vane design.

The 0.8 G force against the left side of the vertical control surface induced an abrupt, left yaw of 10 degrees/sec., into an instantaneous Dutch roll, with a left bank through 25 degrees and a pitch down to -30 degrees.!

Although the pilot used full opposite aileron to counter the radical roll to the left, he did not have the use of right rudder to assist the aileron in countering the left bank.! He had no control of the rudder.!

ATC released AA 587 15 seconds early and then turned him inside the path of the departing heavy B747. AA 587 made a perfect join-up on a horizontal tornado, the 747's vortex.!

For improved flight safety..

William Sherriff
Capt. AAL (Ret)
Flight Safety Consultant

LEM
13th Jun 2003, 01:58
I'd like to express my doubts about this theory.
I know very little about what actually happened, so my opinion could be complete BS.

But I've got two feelings:

First, that the 747 vortex is not so strong to destroy the tail of an airliner like that.

Sure it can create a problem, but severing an Airbus tail is another matter...

Second : Airbus says the rudder was pilot commanded, and there is no proof this theory is wrong.

I had, years ago, an american instructor, ex Vietnam Phantom pilot, whose obsession was using the rudder instead of the ailerons!
He thought me to react to external disturbances with my feet, to keep the heading perfectly centered and wings level, and flying with him was a real torture, almost continuously kicking the pedals left and right!

Now, I say again, I know nothing about the Airbus captain, but I'm just asking myself if maybe he had the same kind of background of my ex instructor...

It would not be the first time we see pilots doing crazy things.

I apologise if my doubt hurts someone, but since a lot of people died, I think it's worth considering all possibilities.
:(

Lu Zuckerman
13th Jun 2003, 02:21
Airbus states flatly that the rudder was pilot commanded.!

This statement is both self serving and an attempt to cover their collective asses.

:hmm:

avioniker
13th Jun 2003, 03:10
It also completely ignores the possibility of the Vert Stab flexing in the mounts because of impending failure.
Wouldn't this cause differential inputs to the rudder actuators. The pulleys are mounted to the empenage in the fuselage tailcone so they would remain fixed.
If the vert stab were flexing left and right the relative length of the input cables/rods would change. That would cause some pretty wild rudder excursions to my way of thinking.
Just one man's opinion.

wsherif1
13th Jun 2003, 14:44
A Previous Quote:

"The 747 vortex is not so strong to destroy the tail of an airliner like that."

The clockwise rotating vortices from the left wing tip of the 747 can register a force of 300'/sec., according to NASA. The vortex is a horizontal tornado with 200plus mph wind velocities. We know what damages tornados can do on the ground, right.!

Quote:

Airbus says the rudder was pilot commanded, and there is no proof this theory is wrong."

At some point in time the linkages were sheared off their connection with the rudder actuators.! See picture of vertical stabilizer being hauled up off the barge. No linkages attached to the three rudder actuators. The rudder was in pieces and recovered 600 yards from vertical stabilizer.!

No evidence the pilot used right rudder input to counter the steep left bank, along with his full right aileron control input. If the pilot had had rudder control, he may have been able to recover the aircraft.!!!

LEM
13th Jun 2003, 16:11
The vortex is a horizontal tornado with 200plus mph wind velocities. We know what damages tornados can do on the ground, right.!

Right, but houses on the ground don't move.

The airplane was travelling at more or less the same speed.
If you consider the vector diagram which compounds the forward velocity of the aircraft with the sideways velocity of the "tornado", you will immediately see the difference of the resultant velocity: the resultant vector will be set at an angle of at least 45 degrees, thus much less powerful.

No evidence the pilot used right rudder input to counter the steep left bank, along with his full right aileron control input. If the pilot had had rudder control, he may have been able to recover the aircraft.!!!
Maybe, just maybe, the wild reaction severed the tail before developing the steep bank.

vector4fun
13th Jun 2003, 23:56
My thoughts,

Right now, this hour, every hour, every day, every week, somewhere in the world, there are aircraft departing and arriving 4-5 miles behind a preceeding B-747. Now, IF anyone seriously believes the wake turbulence from a JAL 747 can BREAK the structure of a sound airframe 4.3 miles behind, then ALPA, APA, NASA, etc. should be DEMANDING FAA immediately increase the required spacing to ten miles or four minutes.

But that's not likely to happen, is it? Because no majority seriously believes it. If it were so, we'd have airplanes being damaged and/or crashing on a weekly or monthly basis at least. Before the current standard became rule, we had much smaller aircraft operating even closer behind 747's. And while several suffered severe upsets, even crashed, they weren't falling apart in mid air.

The JFK controller DID NOT release AAL587 fifteen seconds early, she released the aircraft with the required 4+ miles in trail. The rule book says either standard may be applied.

I can buy the idea that wake turbulence was a link in the accident chain. But there HAD to be other factors in play.

And for the record, I also believe that to assume a professional pilot was deliberately dancing on the rudders enough to cause a .8 G sideload or more is bordering on insane as well. That's akin to the peak cornering capability of a VERY expensive sports car.

Something more was at work here.....

GlueBall
14th Jun 2003, 02:35
AA Captain John Francis LaVelle is on record as having said that AA587 F/O Sten Molin had kicked the rudder pedals while flying as F/O on the B727 at an earlier time.

"He had excellent flying ability, however, he had one strange tendency: To be very agressive on the rudder pedals!"

Captain LaVelle stated that during a climb out in a B727, while the airplane was dirty with flaps at 5 degrees, Mr. Molin stroked the rudder pedals "1-2-3, about that fast."

Captain LaVelle thought that they had lost an engine. Captain LaVelle asked him what he was doing, and Mr. Molin said that he was leveling the wings due to wake turbulence. Captain LaVelle stated that Mr. Molin had never leveled the wings, and his actions had just created yawing moments on the airplane.

Captain LaVelle thought that Mr. Molin was more aggressive than he needed to be. He said that the B727 was a very stable airplane. "Sten was aggressive in his approach to wake turbulence." :(

LEM
14th Jun 2003, 03:53
ahaaaaa..... very interesting to hear that........:mad:

wsherif1
14th Jun 2003, 16:10
Quotes:

"But houses on the ground dont move."

No, they just disintegrate.!

"We had much smaller aircraft operating even closer behind
747's. And while several suffered severe upsets, even
crashes, they weren't falling apart in the mid air."

The smaller the aircraft the less the inertia.! A heavy Airbus
will not move as readily with the applied force, and will absorb
a large portion of the kinetic energy in the structure.

"The resultant vector will be set at an angle of at least 45
degrees."

The -0.8g, (FDR reading) vector force striking the vertical
stabilizer, broadside, on opposite sides of the vertical fin and
rudder, rapidly, would induce extreme bending forces at the
support junction.!

"Now, if anyone seriously believes the wake turbulence from
JAL 747 can break the structure of a sound airframe 4.3 miles
behind, then ALPA, APA, NASA, etc should be demanding FAA
immediately increase the required spacing."

Yes, If the NTSB had not covered up TWA 800's real cause
AA 587 may not have occurred.!

For improved flight safety.

LEM
14th Jun 2003, 21:29
Wsherif1, what is your opinion regarding FO Molin's habit, as reported by GlueBall?

wsherif1
15th Jun 2003, 01:07
Lem,

Captain LaVelle is only one of the many Captains that have
flown with F/O Sten Moline.! There has been nothing but praise
for Sten's character and abilities as a pilot from them.!

LEM
15th Jun 2003, 01:56
Sure, but even the best person in the world can make just one fatal mistake.

Unless Catain La Velle is lying, what happened is in incredible coincidence (a copilot with that background + a severed tail )!

UNCTUOUS
16th Jun 2003, 12:45
IMHO you could have a sideways hammer-blow from a mature wake vortex - but whether it would be laterally sharp (and hard) enough to cause a rudder movement that would snap one, two or three of its actuators? Hard to say - but I would doubt it.

But then again, you have to ask: "So when did the rudder actuators snap and the rudder separate from the fin? Did actuator breakage and rudder hinge detachment occur simultaneously? Was it whilst the vertical fin was still attached to the fuselage?”. In my lay view it would have to have been - because the detached fin's short fluttering descent to the water would not have provided sufficient force to cause the rudder to wholly detach..

Bit of a mystery really when looked at from the angle of rudder detachment timing and forces required. Maybe Rainman would have a view. Obviously the NTSB would have sucked on this aspect already.


Rainman says in reply
unctuous wrote:
>>Did actuator breakage and rudder hinge detachment occur simultaneously? Was it whilst the vertical fin was still attached to the fuselage?<<

Not only would I say yes, but I also think this is another sign of an undamped oscillation that supports my theory of what happened on this airplane. A standing wave oscillation (wake vortex, in this case) that stimulates a closed-loop control system that has negative phase margin is going to cause larger, additive hinge moments in the rudder control system than if it were not out-of-phase. That's because if the control system is 180 degrees out of phase (directly opposing) the aerodynamic load, then the hinge "feels" the force of the aerodynamic load, and it also feels the force its control system is exerting against it, in its out-of-phase attempt to control the airplane.

As the oscillation continues, each new "peak" reaches a new (higher) peak in surface load and therefore hinge moment. If the control system is truly 180 degrees out of phase, then at the point of fracture, the rudder control system would have been pushing one way (say left) with all its force, and the vertical stabilizer would have been pushing the opposite direction (right) with the full force of the aerodynamic response. Rudder breaks and flies off in one direction, tail-fin breaks and flies off in the other.

I still would like to see someone push a big question to the front of the media on this accident: Airbus, could you please publish the "normal" frequency response of the rudder control system, and your "estimated" frequency response of the same system while subjected to your known "rudder synchronization" failure mode described in the subject AD?

Why is this data not available? If there is no smoking gun in this data, not only will I shut up, but other professionals with my knowledge would also comment on that data, and they could pass their own, independent judgment.

Rainman

wsherif1
17th Jun 2003, 04:10
Forget the rudder.! We are talking about a larger and more
effective flight control surface in the vertical stabilizer.! This
extensive surface area of the fin and rudder, when struck by the
force of the rotating vortices, BROADSIDE, first on one side and
then the other, creates a tremendous shear force at the joint
structure.

At some point in time the 0.3, 0.4, and the 0.8 G forces break
the rudder control linkages to the rudder actuators. The free
floating rudder now moves with the directional changes in wind
shear forces and indicates the forces applied on the tail fin.

The designers, due to the requirement for control in engine out
on take-off, have now come up with a very sensitive weather
vane.! The 0.8 G force against the large fin will induce an
abrupt yaw motion, creating an instantaneous Dutch roll into a
steep left bank and a -30 degree dive attitude. There is no
evidence the pilot had use of right rudder to assist in any
recovery attempt, although he used full right aileron application
to counter the left bank.

LEM
17th Jun 2003, 06:07
We are back at the starting point.


I think it's useless to look for noon at midnight: one of the pilots (guess who?) kicked the rudder both ways, as he had already done in the past, and the fin was severed.

If wake turbulence alone could do that, we'd have a crash everyday.

Period.

Ignition Override
17th Jun 2003, 12:18
For those who simply want to blame the deceased FO/Captain: had the FO or Captain ever before (the accident) experienced serious encounters behind larger, widebody jets? If so, had either pilot made fast inputs on the rudder pedals as a result? Memories of such encounters might be subjective and vague, but helpful.

Unctuous: even if Airbus Inc. never releases the data which aerospace engineers could analyse, could data on the 767, DC-10 or MD-11 be similar to that for an A-300, even if some components are built by different sub-contractors? Are similar size Boeing or MD rudders designed to be much less sensitive at normal climb/approach speeds?

After the USAir 737 crash years ago near Pittsburgh (PIT), PA, didn't Boeing use wake vortex data from 727s to keep the focus away from rudder actuators and yaw dampers? Incidentally, as an FO many years ago we once had a DC-9 on initial climb-out begin to smoothly yaw from side to side, to a large angle, and totally uncommanded while hand flying. Nobody was in front of us and the early morning air in Knoxville (TYS) was cool and smooth. The Captain selected manual rudder, and the yaw disappeared fairly quickly, as we quickly did the decent and approach checklists (plus a PA to minimize concern among the passengers and FA's) while turning onto a downwind for an immediate return and declaring an emergency with tower. He had already experienced this problem years before that on an MD-80. Luckily, at a much higher speed near 300 knots, the "porkchop" rudder would have limited rudder movement to about one inch, if my memory (jet-lagged since yesterday in Schipol) is correct.

My main emphasis is that after any major incident or accident, any manufacturer will strenuosly put the spotlight on any other factor except the possibility that its own aircraft can have a malfunction or need for extra inspections, will it not? Just imagine the potential loss in revenue, whether the manufacturer's civilian sector has been subsidized through massive tax-payer subsidies or not.;)

Flamgat
17th Jun 2003, 16:24
Just for interests sake, on the forces of 747 wingtip vortices. Many years ago as a young fighter pilot, I had the opportunity to escort a 747 SUD in a Mirage F1.

I am a bit rusty on the performance of a F1 now, but just to demonstrate the aileron power of the F1, if I remember correctly at 350kts, lowlevel in an airshow, it had a roll rate in excess of 600º/sec with full aileron application.

I don’t know what speed we were descending, as we were flying fairly close formation on the 747, so was referencing my speed to his, but I’m sure a 747 driver can give us the descent speed in a 747 SUD.

We were descending through about 25,000 ft, when I decided to go and find out what the “hype” was about 747 vortices. I was flying on the right wing, and I slid into the vortice from side. With full aileron application to the right, the vortice still rolled me to the left at the speed of a normal straight roll.

So if anyone still doubts that there is a mini tornado behind a 747 think again. This was at 1 “g” and a fairly high speed, say 250 kts +, and fairly light after a 10 hr flight, and minimal pax. Can you imagine the force after takeoff, heavy and slow. Frightening!

It might have been stupid at the time, but it gave me a healthy respect for vortices, and made me a very careful driver after that, whenever I was behind big birds.

GlueBall
18th Jun 2003, 02:31
Flamgat: I don't think that anyone doubts that significant vortexes are generated by 74s.

What most aviators doubt is that those forces are sufficient to rip off the vertical stab and rudder. It's just a fact that heretofore no airplane has had its tail ripped off from such an event.

HotDog
18th Jun 2003, 11:28
Glueball, I beg to differ. There have been failures of metallic vertical stabilizers before. There is a very interesting article on composite primary structures on:

http://www.abaris.com/Downloads/Newsletter1-v4.pdf

Well worth a read.

used2flyboeing
23rd Jun 2003, 07:02
Just spoke to an AIRBUS marketing guy at PARIS, when asked about the A305 ( A300 replacement ) he went right to the NTSB discussion - he said that the flight recorder showed the pilot stomped the rudder L-R-L / OR R-L-R - extremely sharply, in succession & exceeded the ultimate design load of the vertical "+10" what ever that means .... "ultimate +10". The NTSB will get to the bottom of this though .. In high-B maneuvers - the angle of attack of the vertical contributes greatly to the loading of the vertical ( IE the vertical is producing alot of lift laterally at high-B ) & aparently - stomping the rudder may have exceeded design limits - seems plausible - after all the poor pilots threw off both engines due to the violent roll-yawing - so I understand ... AIRBUS used an all composite design - whereas stodgy, conservative old Boeing uses "hybrid-composite" - IE metallic & composite retention on super critical areas such as this - so Ive read on the internet .. will be interesting to get the final scoop though ..

HotDog
23rd Jun 2003, 07:14
I'm led to believe the Boeing 7A7 is entirely of composite construction, not just the empenage but the whole fuselage as well?

GlueBall
24th Jun 2003, 04:37
HotDog: The "metallic" tail failure in 1985 of the JA 74 was attributed to a faulty repair involving a flawed riveting pattern of the pressure dome. The damage had been caused by a tail scrape from an over-rotation incident 8 years earlier.

The survivable "metallic" tail failure of a B-52D in 1964 was attributed, in part, not soley to severe mountain wave rotors. At the time it was still classified, but the crew had been engaged in experimental high speed low level terrain following maneuvers. Needless to say, during simulated combat maneuvers that airframe's operating limitations had been exceeded. Undetected airframe limitations may be exceeded today, but the airframe or any of its components may not fail for many months or years

Recall when in 1986 Dick Rutan and Jeana Yeager had taken off in the scaled composite Voyager for a nonstop, unrefueled flight around the globe: During the lengthy takeoff roll at Edwards AFB one of the wing tips had dragged on the runway and destroyed its winglet. Sometime after the craft had staggered to altitude, upon the recommendation of Burt Rutan, the builder, Dick had induced sufficient yawing motions to snap off the winglet on the opposite wing in order to regain aerodynamic symmetry.

This was but a small example of demonstrated effects of severe airframe yawing.

In the case of AA587, you'll note that both engines had snapped off during similar severe back and forth rudder induced yawing.
This catastophic failure was not caused by wake turbulence.

used2flyboeing
26th Jun 2003, 04:24
For those of you that are scared of plastic airplanes - opps composite - the new Raytheon Premier/Horizon - whose fuselage is wound like a baseball using filiment & resin - one recently ran off the runway during an RTO or something - it hit some fixed immovable structures & tore the wings & gear off - all about escaped intact - fuselage was intact - weighs less that 1000lbs & is as strong as an anvil ..

Ignition Override
29th Jun 2003, 13:06
used2flyBoeing: Would an Airbus, Boeing or other manufacturer Marketing guy do anything except blame the pilots, no matter what type of incident or tragedy? Somehow, I have my doubts that they would discuss such things from a detached perspective, as NTSB, Transport Canada, British accident board members attempt to do.

used2flyboeing
7th Jul 2003, 23:06
Agreed - hard to argue with the flight recorder though - there are not many A300s left that are hauling people - United to my suprised uses them on red-eyes etc. But most are used for freight - I thought after 911 most of them would be retired - particularly after the FAA mandated a thrust revevser lock update that comprised not only clap-trap ( WSJ had an article onit last year ) - but new expensive nacelle cowls - to cover the clap-trap. But, I guess operators want to continue operating these old birds..

LEM
26th Oct 2004, 19:26
WASHINGTON (AP) -- If the pilot flying American Airlines Flight 587 had taken his foot off the rudder pedal, the jetliner's tail wouldn't have broken off, the plane wouldn't have plunged into a New York City neighborhood and 265 people wouldn't have died on Nov. 12, 2001.

Airdisaster.com, today....