NTSB orders inspection of another Airbus

[N380UA]

WASHINGTON (CNN) -- Federal investigators have ordered the removal and inspection of the rudder and tail fin of an American Airlines Airbus A300-600 involved in an in-flight incident in 1997.

<a href="http://www.cnn.com/2002/US/02/25/ntsb.flight.587/index.html" target="_blank">Read all about it.</a>

It seems as though the structure of the vertical stab may be compromised by wrong handling and apparent poor airmanship. Any 300 driver that can confirm that?

[Notso Fantastic]

I'm wondering where the 'poor airmanship' references have crept in since the JFK accident. I am taught in 'recovery from unusual attitudes' to use aggressive and large inputs if needed. To this end, I have recently in the simulator been using full aileron in a 747 to so recover from such attitudes. Maybe these pilots did use large and aggressive rudder inputs. Is this so wrong? Is it a 'pilot error' or insufficient design integrity when the rudders apparently fly off? Every time I see a reference to pilot error with this sort of thing, I'm more inclined to see it as poor design/testing.

[ironbutt57]

No crew error involved..IMHO...

[HOVIS]

If the bl**dy structure cannot cope with full scale deflection of control surfaces then what the hell is the point in having such a degree of movement?. .If this was Concorde it would have been grounded by now, but there again the 737 has a rudder system that is 100% safe..... Does it not?

[apfds]

I presume all pilots accept the fact that if one pulls to much 'G ' then the wings may fall off.. .Why then is it so difficult to accept that the vertical stabilizer can also be overstressed.

[M14P]

Oh dear!

A wholesale misunderstanding of both large aircraft and swept surface aerodynamics.

Sudden control deflections particularly if violently checked could overstress the fin. That is true of virtually any design with a swept fin due to its ability to create large angles of attack. Add to that a flight regime which is at the upper end of the speed range for full (i.e. unlimited) rudder travel and you have a recipe for overstressing the fin.

Just take a moment (no pun intended) to think about the amount of force that the fin would have to generate to produce the rates of yaw and controllability to acheive certification for a 100+ ton airliner. Then consider that the surface is attached only at one end and must deal with non-linear forces through twisted axis thru a speed range of 70 to 400 knots IAS and MNs up to .85+. Now tell me how you would help to protect the surface from overly agressive control inputs?

I'll tell you how: A Fly By Wire protected envelope. The A300 is not a FBW aircraft and the A320 only has conventional rudder limiters with a very fancy yaw damper. One day, with development and engineering aircraft will become far more difficult to break.

Fin separation at very high sideslip angles is not new (e.g. Convair B-58) and therefore we should not be surprised if it occurs. Advanced Manouevre Programmes in the 'wake' of some 737 accidents taught pilots more aggressive control inputs including use of rudder to improve roll rates. However well intentioned that this was it seems that the dangers of imprudent rudder application were not highlighted enough.

Conclusion: Unusual attitudes / jet upsets, use aileron/spoiler up to its full authority then carefully augment with rudder.

[Wino]

M14 As only an A300 has had the tail sheered off by rudder inputs I would argue that other than the A300 aircraft structures have been sufficiently strong to handle significant mishandling over the 40 years that jets have been flying. Furthermore as it may very well . .have been a computer problem that caused the inputs that broke the tail of the A300, more computers may not be the answer but LESS.

But also of interest is that apparently AA took an aircraft and put it up on jacks with the engines running and all kinds of things hooked up to the pitostatic system and IRSs to trick the aircraft into thinking it was in flight and some VERY interesting things have come to light.

The Load limiter on the A300 we have found works by limiting travel of the rudder pedals, it comes up from the floor limiting how far you can push the pedals. at 250 knots total travel of the pedals turned out to be 1 inch in each direction. For some reason the travel gets further again later as you go faster.

That was interesting, what was even more interesting was at 250 knots breakout force was only about 5-10 pounds less than that required to get FULL travel of the rudder. The way the aircraft performed its almost like an all or nothing system. The Sim that AA has performs nothing like that (I know).

We have just received warnings from the company about it. They did the same thing with a 757/767 and the rudder turned out to be far more predictable. The fly in the ointment of this test is that there was no airload on the fin that I know of so maybe this whole thing doesn't matter, but just the way the load limiter works (by blocking movement from the floor up rather than ratioing it) is so different from the sim and Airbus's AOM that it was eye opening.

Cheers. .Wino

[ 26 February 2002: Message edited by: Wino ]

[ 26 February 2002: Message edited by: Wino ]</p>

[NigelOnDraft]

2 thoughts... (from a B757/767 P2)

1. UP recovery: Whilst I have done this in the sim, it has concentrated on aileron / elevator. Pretty impossible to overstress on aileron (although care needed if combined with 'g', hence roll then pull), elevator natural / known limit of 'g'. Never any hint of using rudder, which I know should be treated with care...

2. Hints over the AA accident has referred to a large rudder deflection one way, very soon followed by a similar but opposite deflection the other - the reversal causing a larger problem than just a single large deflection. No proof that this was due to crew inputs (but a possibility). A reminder to us, without (public) accusations at either crew or aircraft (for now?)

NoD

[Flight Detent]

Just a simple question -. .Can anybody remember an Airbus incident/accident report from Airbus that didn't put the cause as 'Pilot error', ever?. .Cheers.

[John Farley]

NoD

Your 2 is so right

[Captain104]

NoD. .You just hit the nail. It sounds so simple and reasonable. But wait what comes next. <img src="wink.gif" border="0">