Maneuvering airspeed for Airliners
 

I was reading an article on an airliner crash in the May 2003 issue of the US magazine FLYING. The article by author Les Abend was called Beyond the Rubble; The crash of AA Flight 587. That crash of an Airbus A300-600 occurred at New York in November 2001.
The aircraft had taken off behind a JAL Boeing 747 and a few minutes later the tail of the Airbus separated from the fuselage causing the crew to lose control and crash. The NTSB determined that the copilot was the pilot handling when the tail separation occurred.

From the digital flight data recorder (DFDR) it was determined that the rudder may have moved to its full limits, back and forth in quick succession, in a very brief period of time. The writer of the article stated that Maneuvering speed, if that is the true definition, is not published in airline operating operating manuals. Operation Manual sections publish never-exceed speeds (Mmo) but not maneuvering speed.. Flight 587 never even came close to a limitation speed. So why did the tail exceed its ultimate load?

The Airbus 300-600 requires very little pedal movement to affect large movements of the rudder compared to other hydraulically powered transport-category aircraft. For example, only 1.5 inches of pedal travel will move the rudder to full deflection at an airspeed of 250 knots.. That is very close to the airspeed that Flight 587 attained while it encountered the wake turbulence of the preceding Boing 747. At that airspeed, rudder pedal movements requires only 37.5 pounds of applied force. At an airspeed of 330 knots, rudder pedal movement of merely one inch and a force of 30 pounds is all that is required for full deflection. The writer postulates that if the copilot of Flight 587 actually overcontrolled the rudder, do you think he would have done so if he knew there was a possibility that the tail could depart the aircraft?

The figures for full rudder travel quote above are for the A300-600. I wonder if the same figures could be valid for todays Boeing 737 or fly by wire types? Either way, it is doubtful if type rating engineering courses even mention rudder travel versus rudder pedal travel. It certainly was never covered on the Boeing 737 type rating course I did many years ago and probably not covered in current simulator training. Maybe it should be?

That event is a long and complex story with many tentacles.

From the findings https://www.faa.gov/sites/faa.gov/fi...7_findings.pdf
As for the Airbus FBW types....
https://safetyfirst.airbus.com/use-of-rudder/
https://safetyfirst.airbus.com/wake-vortices/

It seems that this "foot off" training is not just needed in the airliner world. Look at the data traces here https://skybrary.aero/sites/default/...shelf/4040.pdf

FAR Design Manoeuvring Speed is addressed at eCFR :: 14 CFR 25.335 -- Design airspeeds. (FAR 25.335)

The relevant FTG is at AC 25-7D, Flight Test Guide for Certification of Transport Category Airplanes

and, for the small brigade, a read through the relevant bits of the FTG at AC 23-8C - Flight Test Guide for Certification of Part 23 Airports is time usefully spent over a coffee or ten.

I didn't check the EU story for this post but it should be somewhat similar.

The main problem (put into glaring relief post AA 587) relates to general misunderstanding of what Va is. The NTSB report at

In-Flight Separation of Vertical Stabilizer American Airlines Flight 587 Airbus Industrie A300-605R, N14053 Belle Harbor, New York November 12, 2001

should be mandatory reading for all aircraft pilots, both small and large certifications.

Post AA 587 the Industry undertook a significant training program to "unlearn" the general pilot misconceptions as to what Va was and was not. This is still ongoing, and slow, especially with the continuing retelling of the misconceptions in new pilot training. Being involved in such things in recent years, I do my bit but my voice is a voice in the wilderness, as it were. Others with a certification background do likewise but, sometimes, it all looks like a losing battle. On the small aircraft front, the problem probably will slowly go away as we see more aircraft move to the use of Vo in lieu of Va.

I'm guessing that's a rhetorical question ...

The B 757/767 had a system called rudder ratio system. This reduced the rudder angle for a given pedal input as a function of airspeed, thus reducing fin loads.
I think somewhere in D Davies' book, "Handling the Big Jets" he emphasises in upset recovery to "stay off the rudder", he didn't on the A 300 accident.

I seem to remember from the report that the failure was not caused by a single full application of the rudder but rather it was the rapid cycling of the rudder between full deflection in both directions several times which was outside the stress calculations assumptions.

the failure was not caused by a single full application of the rudder

Indeed.

The story is that, at Va, a control surface and attachments can sustain a single full application (or, if appropriate, a single checked input to the maximum control load value) and then a relatively gentle release. This is one of the reasons I suggest that the full report should be mandatory reading for pilots. Va, unfortunately, is still very much misunderstood by a significant proportion of pilots

The level of physics pilots are equipped with from training usually stops at the diagram below. Better diagrams include lines of gust values from which you can read the G-load at different airspeeds. Whether the average line pilot these days can even explain what this is, I am not so sure... The FCOM or QRH has a list of turbulence penetration speed, hopefully they'll have that memorized at least. As a last resort the Airbus engineers put a load factor limit in the side stick :O Except for engine failures, the rudder is simply not being used in training or line flying these days. The yaw damper does a fantastic job. The only time I'd hoped to see a bit more rudder activity is during crosswind landings, but not even then people seem bothered anymore...

https://cimg5.ibsrv.net/gimg/pprune....7ee6dadde4.png

The NTSB report discusses the designs used on various types. It starts on page 26 (page 40 on the PDF). If you scroll down a couple of pages, there's a table of figures.

The 737 doesn't have a rudder ratio changer like the later Boeing types. The ratio of rudder travel to pedal travel should be constant. It has what the NTSB called a force limit system. There's no specific rudder travel limit for a given airspeed, instead there's a limit on the force applied by the actuator. That means that in theory, it should be resistant, if not immune, to the type of failure that happened with AA587. If you put it in a left sideslip and then slam the rudder to the right, the rudder simply wouldn't travel as far as it usually would. The ultimate structural load wouldn't be exceeded. Apparently the DC-10 and MD-11 also use a similar design.

IIRC a contributing factor was partial delamination of the vertical stabilizer caused by leaking hydraulic fluid.

One minor gotcha which is part of the pilot understanding problem. Ref post #8, the reference to manoeuvring speed really is to the MINIMUM Va. Va, depending on OEM design decisions, can go quite a bit higher, typically not above Vc. Vo fixes this problem as it limits the idea to the minimum Va speed.

The funny thing is, that Boeing actually published a TIB in the aftermath basically saying: if you act as stupid as that pilot the same thing will happen (multiple repeated full travel inputs will lead to structural failure). Back then i was still on the 737-300/500, but it was still in the FCOM once we moved to the NG.

I get the occasional contingency brief that mentions manual use of rudder in the cruise with an engine failure. I point out that a) even with full thrust on one side and nothing on the other, the residual yawing moment is significantly smaller compared with V2 at sea level, b) the AP can cope with it until it’s trimmed out and c) most of the time TAC will do its thing. A heavy boot on the pedals is unlikely to make the situation more manageable, especially if all you’ve practiced is V1 cuts.

Doing acceptance flights was a very valuable tool in this regard - regarding one thrust lever at altitude and then back up to watch spool up times, one simply left the AP in to handle it, which on the 737 didn’t have a rudder channel.

The 737 can have a rudder channel. But even then it is only active during approach below 1.500ft and during go around until another lateral mode is engaged. And most airlines don't order it, although it allows CAT IIIb approaches.

Whilst I’m aware it can, this was the classic which I believe was not available with that option.

Flashbacks to when Boeing tried to blame the dead 737 crew of US427 for incorrect actions in wake turbulence when it was the uncommanded rudder hard over.

Often people confuse the "design manoeuvring speed Va" which is a FAR/CS-25 structure certification definition with "flap manoeuvring speed" which is a OEM definition of an airspeed which allows for "safe" manoeuring (usually 25º of bank with a 15º overshoot allowed) without stalling. The former is a maximum speed whereas the latter is a minimum.

In terms of the design manoeuvring speed, Va, I don't believe OEM's of large jets usually informs the pilots of their value because they don't really need to know. Same as Vc, Vd and other structural airspeeds. We stick to our operational values, VMO, VNE and so on, these all have healthy margins baked in versus the structural airspeeds for all flight conditions.

Below Va, an airplane will withstand a single full application of a control surface, but not necessarily a repeated, oscillating application which is a completely different structural problem. Each cycle of control application increases the loads on the fin because the sideslip angle becomes larger and larger with each swing. While the training the F/O of AA587 underwent maybe didn't highlight the dangers of multiple, repeated and alternating control input applications, I also don't believe instructors were actively training students to react this way. It feels like the overreacting tendency of this F/O fell through the cracks without anyone noticing or bringing it to his attention to correct it.

Afaik every Boeing limitations chapter explains to "avoid rapid and large alternating control inputs, especially in combination with large changes in pitch, roll or yaw as they may result in structural failure at any speed". Important detail: avoiding "rapid control reversals" is valid in _any_ axis: yaw, but also pitch and roll.

The rudder limiting/reducing discussion isn't really important here, as the AA upset recovery focused on rudder _at low speed_. Which was countered by Airbus that at low speed the rudder is simply created for max effectiveness: engine failure at low airspeeds. The "rumors" were the AA upset recovery program was designed and influenced by people with a heavy military background, flying aircraft with high thrust-to-weight ratio's. This is actually mentioned in a footnote in the report: "The Airbus chief test pilot indicated that fighter pilots are accustomed to using the rudder for evasive maneuvers when flying not far from stall speed but that large airplanes are not similar to fighter airplanes."

The report is indeed a must read for anything related to upset recovery.

"It is important that the first actions for recovering from an airplane upset be correct and timely. Exaggerated control inputs through reflex responses must be avoided. It is worth repeating that inappropriate control inputs during one upset recovery can lead to a different upset situation."

https://www.ntsb.gov/investigations/...ts/AAR0404.pdf

As far as the original question is concerned: rudder (or any other flight control) manipulation should be a conscious act, not just "guesswork to get the thing moving as quickly as possible". The modern jetliner (and with that I even dare to include anything with a yaw damper) is controlled through dedicated controlled inputs. Those forces for those inputs are fairly similar over a wide range of jets and I honestly think a pilot has enough time to adapt through a type rating course and line training.

The NTSB said otherwise. Read my post #2. That’s from the report.

there was plenty of unofficial commentary outside the report confirming that aggressive rudder use was being actively taught by an over zealous ex military instructor.

Again, the biggest problem with Va is the general misunderstanding of what it is about. The general understanding is in line with the relatively new term, Vo, and, I presume, in due course, we will all cease to worry about Va which is an engineering thing, rather than operational.

Often people confuse the "design manoeuvring speed Va" which is a FAR/CS-25 structure certification definition with "flap manoeuvring speed"

Va applies to light aircraft also.

Below Va, an airplane will withstand a single full application of a control surface

Caveat. If a full application would put the aircraft above the limit load factor, then the input has to be a checked manoeuvre.

aggressive riders use was being actively taught by an over zealous ex military instructor.

The upset recovery training program was brought to account post accident and subsequently amended. I guess that it is very difficult to train line pilots in the ways of certification test pilots. Mostly things work OK as we don't have aircraft falling out of the sky with no fins on a day by day basis ?

It’s also important to recognize that after AA587, in accordance with NTSB recommendations, the FAA actually did change the legal definition of how “Maneuvering Speed” is used in the AFM, per 14 CFR 25.1583. It is unfair to judge the pilot’s actions at the time of the accident against the current regulation and information provided to he flight crew today.

https://www.federalregister.gov/docu...tion-statement

Some good discussion on the addition of Vo to Part 23 as well.

FARs
 

I think it more a case of trying to get the correct story through to the flying folk. The definition and associated conditions go back at least to the 50s.

The problem is more a case of the training side of the Industry not making sure that the story they give to the up and coming trainees is correct. Va has been one of my pet peeves for ever and a day.