Don’t tell Evan

[3WE]

And what now is the thing that Evan shouldn't know?

When the A-330 crashed in New York back in 2001, after the tail broke off...

Evan feels that the pilot, in response to poor training and screening, deliberately slammed the rudder back and forth, repeatedly, because he specifically wanted to be a cowboy, improvising pilot and wanted to use rudder BECAUSE it goes against the Airbus A300B4-605R FCOMPOHQRH.

3BS feels that the pilot felt some wake turbulence and made some modest, appropriate rudder inputs (Maybe 1.25" and 10 lbs...not_unlike the landing fake cub in this video). He used the rudder because that's a kinda fundamental control to manage yaw in lots of airplanes in lots of circumstances. Unfortunately, his modest light inputs were grossly amplified to the full, available rudder deflection by a poorly designed airplane, resulting in an unfortunate, systematic oscillation which broke the cheap-ass composite tail off of the airplane.

The truth may be in the middle, but I don't feel the same need as Evan to make the pilot out as an idiot.

[Gabriel]

When the A-330 crashed in New York back in 2001, after the tail broke off...

A300. And the difference is not minor. This accident would likely not have happened in an A330.

[B]He used the rudder because that's a kinda fundamental control to manage yaw

Well, the problem is that apparently (and yes this is a stretch) he did not use rudder because to control yaw, but he thought that wake turbulence could flip the plane so he had to use upset recovery procedures and he attempted to use rudder to control roll rather than yaw (at least initially). Which goes back to...

the pilot, in response to poor training...

[Evan]

The truth may be in the middle, but I don't feel the same need as Evan to make the pilot out as an idiot.

Not an idiot, perhaps, but a pilot trained for wake turbulence upset recovery on a very different DC-9. Different because the smaller wingspan of a DC-9 allows for the possibility of severe wake upset attitudes where both ailerons can be inside the vortex and the only means of addressing roll at such extreme attitudes may be rudder. Even so, not with rapid reversals. When pilots who had previously flown with this pilot observed his very active use of rudder, they later remarked to investigators that it was disturbing to watch. Like you, this pilot failed to understand the role of rudder on large transport category airplanes and this got a lot of people killed. The sensitivity of the rudder pedals is not the fatal factor, it's the rudder reversal behavior that caused the fin to depart. When you use rudder to induce roll, there is a delayed effect. Therefore it becomes very easy to overcontrol and get into an escalating PIO situation. It's a terrible substitute for ailerons. At least one, and probably both, ailerons remained effective on the A300 during the event. But even they weren't needed. The prudent thing to do on an A300 (or any aircraft of that size) is let the airplane's lateral stability naturally correct itself. Maybe some gentle aileron but not rudder. The pilot responsible is now dead and will never learn from this quite blatantly demonstrative lesson. What is your excuse 3WE?

[Not_Karl]

Ban ALL rudders.

[Evan]

This accident would likely not have happened in an A330.

What makes you say that?

[BoeingBobby]

Had he not flown that little Scarebus so close behind that 74...

[Gabriel]

What makes you say that?

All these airplanes need a very powerful rudder to control the critical engine failure with max thrust in the other engine at Vmcg which necessarily needs to be lower (by a small margin) than the minimum possible V1 (imagine a very lightweight take-off from a very short runway). At the same time, at the structural maneuver speed Va they need to withstand a sudden maximum rudder deflection to the point of max overswing and a sodden rudder centering from there. Those two requirement are kind of incompatible so the solution is to make said rudder deflection variable, so you get the real max max during take-off but the max travel reduces as speed increases.

In the A300, the limiting is done in the pedals themselves maintaining the pedal-to-rudder displacement ratio (called just rudder ratio) constant, which means that at higher speed (but still below Va) you have much less rudder pedal travel available from neutral to the "local" max deflection. Less deflection to reach the local max also means less force to reach the local max, and because it takes some force just to even start moving the pedals, the force needed to apply the 2"of so of max pedal was not so much greater than the force needed to even start moving them.

The A330, as mot planes today, have instead a variable rudder ratio, where you always need max deflection (and high force if going fast) to apply the local max rudder. That would have made a full rudder PIO, requiring much more pedal deflection and force, physically much more difficult to accomplish, and hence much less likely.

[Evan]

All these airplanes need a very powerful rudder to control the critical engine failure with max thrust in the other engine at Vmcg which necessarily needs to be lower (by a small margin) than the minimum possible V1 (imagine a very lightweight take-off from a very short runway). At the same time, at the structural maneuver speed Va they need to withstand a sudden maximum rudder deflection to the point of max overswing and a sodden rudder centering from there. Those two requirement are kind of incompatible so the solution is to make said rudder deflection variable, so you get the real max max during take-off but the max travel reduces as speed increases.

In the A300, the limiting is done in the pedals themselves maintaining the pedal-to-rudder displacement ratio (called just rudder ratio) constant, which means that at higher speed (but still below Va) you have much less rudder pedal travel available from neutral to the "local" max deflection. Less deflection to reach the local max also means less force to reach the local max, and because it takes some force just to even start moving the pedals, the force needed to apply the 2"of so of max pedal was not so much greater than the force needed to even start moving them.

The A330, as mot planes today, have instead a variable rudder ratio, where you always need max deflection (and high force if going fast) to apply the local max rudder. That would have made a full rudder PIO, requiring much more pedal deflection and force, physically much more difficult to accomplish, and hence much less likely.

Understood, but again, the fatal factor here was not in deflecting the rudder to max deflection and centering it from there. No matter the force applied by the pilot, this would not have overstressed the airframe. Rudder travel is limited in either application. The fatal factor was moving the rudder from left max overswing to right max overswing. What reason would any pilot ever have to do this on a large aircraft? This pilot was trying to use the rudder improperly as a flight control to address roll instability, which induces roll responses with a delayed effect that is a perfect recipe for PIO, large rudder inputs to both sides and structural failure. It's not a primary flight control. It's a compensatory flight control.

Input rudder as needed. Remove rudder as needed.

[brianw999]

Uuuummmm…….. talking of rudder use…..

https://youtu.be/roS6oFjCDhc

[Gabriel]

Understood, but again, the fatal factor here was not in deflecting the rudder to max deflection and centering it from there. No matter the force applied by the pilot, this would not have overstressed the airframe. Rudder travel is limited in either application. The fatal factor was moving the rudder from left max overswing to right max overswing.

Right, but you are not going to reach max overswing using only partial rudder. That is why this accident was unlikely to happen in an A330 because with its rudder pedals design, it would have made a full rudder PIO, requiring much more pedal deflection and force, physically much more difficult to accomplish, and hence much less likely. PIO. I said PIO. I didn't say single rudder input.

What reason would any pilot ever have to do this on a large aircraft? This pilot was trying to use the rudder improperly as a flight control to address roll instability, which induces roll responses with a delayed effect that is a perfect recipe for PIO, large rudder inputs to both sides and structural failure. It's not a primary flight control. It's a compensatory flight control.

Input rudder as needed. Remove rudder as needed.

What? So do you think that the pilot was intentionally applying full opposite rudder at the point of max overswing, because he didn't understand rudders? Do you understand PIO?
It doesn't need yaw-roll coupling. You have PIO just in pitch. You have PIO just in roll. And this PIO was mostly yaw with minimum roll. Like the PIO that happens in cars.

Define "as needed" in "Input rudder as needed".

[Evan]

Right, but you are not going to reach max overswing using only partial rudder. That is why this accident was unlikely to happen in an A330 because with its rudder pedals design, it would have made a full rudder PIO, requiring much more pedal deflection and force, physically much more difficult to accomplish, and hence much less likely. PIO. I said PIO. I didn't say single rudder input.

I'm really exhausted on this subject. I'll let the investigation speak for me:

In the following section, a conclusion is reached through syllogism, based on the judgment that there is PIO and that this phenomenon is attributable to the sensitivity of the airplane's flight controls. This short cut fails to take into account induced roll, characteristic of rudder control, that is common to all transport airplanes in this category, which generates a time delay between the pilot's control input and the response perceived. The Airbus simulation on the relation between PIO on the roll axis and the sensitivity of the controls appears to show that even a light rudder pedal input may create a divergent output. This system remains stable simply by using the control wheel with high gain. As far as we know, no studies have been undertaken on PIO on the yaw axis since the rudder is not a primary flight control.

Extremely high external loads were developed on the accident aircraft due to the repetitive, alternating, aggressive stop-to-stop pilot inputs on rudder pedals.

Of the 6 pilots who flew the AA A300 simulation, which was recovery from a high-bank upset event (90-114deg roll, which did not occur on the A300, nor would they), 5 found that the best strategy to recover was either large wheel inputs and no rudder or large wheel inputs and a small rudder input.

Again and again, the exhaustive report points out that the use of rudder to control yaw creates sideslip forces that can lead to loss of control and structural failure.

Define "as needed" in "Input rudder as needed".

On civil transport category airplane, the rudder pedal is more a zeroing flight control to compensate for any yaw asymmetry than a primary flight control to create yaw asymmetry as it is on some military fighter aircraft.

On large transport category aircraft such as the A300B4-605R, roll control authority is adequate, even in the case where upset recovery techniques must be applied. Furthermore, rudder doublets (full stop-to-stop pedal deflections such as those observed in this accident) are not recognized design conditions, nor is there ever an operational need for them in transport category aircraft.

See the FCOM attachment, (Appendix N of the factual report) which pre-dates the accident ----->

[Evan]

Had he not flown that little Scarebus so close behind that 74...

He was downwind and below the 74. The wake travelled laterally with the ambient wind and descended into the path of AA587. Which was by no means little and would have rode it out bravely with a competent pilot at the helm.

[Evan]

A300. And the difference is not minor. This accident would likely not have happened in an A330.

I thought you were referring to the yaw damper logic. On the A300:

The yaw damper actuator signals are added to those of the pilot, up to the maximum travel allowed by the variable stop lever. Due to this logic, the pilot authority is always greater than the yaw damper authority.

I'm not sure if this is true on the A330.

[Gabriel]

Ok, Evan. You win. This accident would have also happened had the rudder pedals required 10 inch of travel and 50 pounds to apply max deflection. The fact that a minimum rudder displacement with a additional minimum force of top what is required to barely start moving the rudder is enough to effect full rudder input has nothing to do either with this accident or with the propensity of PIO.

[Evan]

Ok, Evan. You win. This accident would have also happened had the rudder pedals required 10 inch of travel and 50 pounds to apply max deflection. The fact that a minimum rudder displacement with a additional minimum force of top what is required to barely start moving the rudder is enough to effect full rudder input has nothing to do either with this accident or with the propensity of PIO.

It absolutely has something to do with it. The more sensitive controls exacerbate the situation, but it is not the causative factor to focus on. Again, if the pedal required only the slightest touch and was applied in one direction and then removed, there would be no structural failure. Better yet if the rudder was not touched at all. I also believe this pilot would have broken the A330 or the 747. It just might have required more work.

Anyway, I’m not out to win, I’m just pointing out the vital lesson the investigation teaches about the role and proper use of rudder on large transport category airplanes. It isn’t useful to propose that this lesson is less relevant to another type because the controls are more forgiving. It’s unforgivable either way.