Air France 447 transcript

The friendliest place on the web for anyone who enjoys boating.
If you have answers, please help by responding to the unanswered posts.
Excellent post, with much to learn from.
 
1. It's amazing to me that this professional pilot had such a poor understanding of how to fly his airplane! I'm afraid some pilots have learned to become systems managers, not intuitive pilots.

2. I believe all planes should provide tactile feedback to all system flight controls so fellow pilots can see and feel what's happening on the aircraft.
 
1. It's amazing to me that this professional pilot had such a poor understanding of how to fly his airplane! I'm afraid some pilots have learned to become systems managers, not intuitive pilots.

2. I believe all planes should provide tactile feedback to all system flight controls so fellow pilots can see and feel what's happening on the aircraft.

Another question you have to ask yourself is would this have happened if it was a Boeing? I will answer by saying...not likely. It is easy to just blame the pilots and while I think some of the conclusions in this article are good...one is that this pilot likely did exactly how he was trained. Airbuses have "envelop protection"...and this guy thought he was within the envelop(being protected by the computer) the whole way to impact. And they were completely baffled the whole way down because the aircraft wasn't doing what they thought it should be doing. One thing I have against Airbuses is that they are simply flying computers...garbage in, garbage out. And this guy was feeding it garbage all the way to the surface of the ocean. A Boeing gives you a lot more clues to what is going on with the aircraft.....and at the very least, you can see what the other guy is doing just by the position of the yoke. Anyway, I am not one to armchair quarterback the crew. It is a dangerous/arrogant pilot that thinks it could never happen to him. I think some major contributing factors are Airbus philosophy and the training from the airline. Do they ever question Airbus design/philosophy? Not really other than the control sticks are not really visible by the other pilot.
 
Agreed. That's the tactile feedback of the flight controls I was referring to. Besides the fact you mentioned of being out of sight, the Airbus sidesticks do not deflect with cross-pilot input. I'm told they also do not 'load up" with control force as the airplane slows. Not sure if this is accurate or not, but if so, it doesn't sound much different than an inexpensive computer joystick from the 1990s.
 
The failure in the system is the simulator training "approach to stall", as a practiced drill.
 
I'm told they also do not 'load up" with control force as the airplane slows.

Don't know either but I think I would prefer the stick to "load up" as speed increases. Low(er) stick forces at low speeds, high stick forces at high speeds to provide feedback and reduce overcontrolling.
 
Not only tactile to pilot in command, but to the first officer as well. Asynchronous controls without tactile feedback is troubling and a key difference in the Boeing design philosophy. I don't think we'll ever be changing ours. It's the reason we've stayed with conventional (cross-connected) yokes.
 
1. It's amazing to me that this professional pilot had such a poor understanding of how to fly his airplane! I'm afraid some pilots have learned to become systems managers, not intuitive pilots.
I haven't flown a plane since about 1988-89 but this report really upsets me. Pilots who are just monitoring computer generated data and don't really know how to recognize/recover from a stall. Where is all this heading? A tail falling off an Airbus in new York, another Airbus falling out of the sky over the Atlantic. I don't even want to go near that plane!:mad:
 
Normal flight controls (non-Airbus) become sluggish at slow speed and fluttering flight control surfaces and wing buffet can provide a tactile clue as one approaches a stall. Aircraft are trimmed for the speed of flight to minimize the control forces. This trim is adjusted by the pilot or autopilot as the desired speed changes.

If one is flying slower than desired, the nose wants to drop and it requires excessive back pressure to hold the nose up and the flight controls are sluggish. Conversely, if the plane is flying out of trim faster than desired, the nose wants to lift and excessive forward pressure is required to hold the nose down and flight controls are highly effective. Pitch trim adjustments relieve these nose-up/down forces.
 
With over 12,000 hours in Gulfstream II, III, IV and Vs, I think the feel of the controls given to the pilot are the best design. Your body always knew what your aircraft was doing. I found to many new pilots relied completely on the auto pilot and FMS. The only time they would do it by hand was to taxi. We would click off the A/P crossing the pond and never spill the drinks. Kept us sharp to hand fly to cruise altitude. To many robots in the cockpit.
 
Airbusts are built so 3rd world "pilots" do not kill folks with minimum training..

A new Paris air show is coming up, any bets if for the 3rd time a new Airbust crashes in front of the crowd , with test pilots performing?
 
Don't know either but I think I would prefer the stick to "load up" as speed increases. Low(er) stick forces at low speeds, high stick forces at high speeds to provide feedback and reduce overcontrolling.

Rick, you won't overcontrol it....the computer won't let you! You can slam the stick back and forth at speed and the aircraft will gently rock it's wings....okay maybe not gently but it will only roll at the rate limited by the computer.
 
I haven't flown a plane since about 1988-89 but this report really upsets me. Pilots who are just monitoring computer generated data and don't really know how to recognize/recover from a stall. Where is all this heading? A tail falling off an Airbus in new York, another Airbus falling out of the sky over the Atlantic. I don't even want to go near that plane!:mad:

Those two crashes are unrelated. The NYC crash was an A300(or maybe 310) but is not fly by wire. They hit turbulence and for some reason one of the pilots used the rudders to counteract the turbulence and over stressed the rudder.

Because of the computers on the Fly-by-wire Airbuses, some pilots use a "last ditch effort" mentality of pulling back on the stick and full power on the engines. The "envelop protection" of the computers will keep the airplane from stalling and you should climb. The problem here was that the pitot tubes iced over and the computers reverted to "alternate law" which means that envelop protection was lost. They stalled the airplane. The pitots unfroze and the normal law returned but they were already outside the envelop...and the computer could not protect them and the pilots did nothing to help it. It took a moment for the captain to figure it out but it was too late....
 
Airbust first 350 to fly Friday , for how long?

Till someone attempts to actually fly the aircraft in stead of just making computer suggestions the computer likes.
 
I don't get why both copilots were in the seats and the Jr one was flying the plane. Shouldn't the more senior person be in charge?

Sent from my iPhone using Trawler
 
Those two crashes are unrelated. The NYC crash was an A300(or maybe 310) but is not fly by wire. They hit turbulence and for some reason one of the pilots used the rudders to counteract the turbulence and over stressed the rudder. Unrelated? Aren't those two planes built by the same company? I wonder what forces an F-35 sees on its rudder.
If it's an Air Bus. I won't go near it!
 
I don't get why both copilots were in the seats and the Jr one was flying the plane. Shouldn't the more senior person be in charge?

Sent from my iPhone using Trawler

I believe this was an 'augmented crew'. That means they have an extra pilot on board for very long flights to allow a pilot to rest/sleep in a separate sleep area during the flight. When the Captain sleeps, the 2 other pilots control the airplane. One is assigned Capt duties during this period, but the official Captain retains ultimate authority when not in rest. All pilots assigned Captain duties must be Captain qualified, so it's not like some junior inexperienced pilot is given the command of the aircraft.
 
I believe this was an 'augmented crew'. That means they have an extra pilot on board for very long flights to allow a pilot to rest/sleep in a separate sleep area during the flight. When the Captain sleeps, the 2 other pilots control the airplane. One is assigned Capt duties during this period, but the official Captain retains ultimate authority when not in rest. All pilots assigned Captain duties must be Captain qualified, so it's not like some junior inexperienced pilot is given the command of the aircraft.

True. The way it works at our company is your "awarded" position determines the chain of command. The Captain is PIC. The first officer is second in command(and PIC when CA is not on flight deck) and then the IRO(international Relief Officer). Usually the FO is more senior than the IRO....but not always.
 
Last edited:
If it's an Air Bus. I won't go near it!

Walt, my point was that the crashes and their cause and the aircraft involved were very different...regardless of the fact that it was the same builder. The Air France crash was a fully functioning airplane that was flown into the ocean by the crew. The AA crash was structural failure and subsequent loss of control.

But I agree. My personal opinion is that Boeing makes a better airplane.
 
A good friend and old Navy aviator put it more succinctly. He said, "don't fly with Frogs".:hide:
 
And I thought TSA was reason enough to avoid airports!
 
I do agree with Boeing being a better aircraft. I know a few United pilots who call them selfs "BUNKIES" . Most of them really do not want to fly. Just do the minimum to keep their job. Many times they run out of landing currency and have to go get sim time to be legal.How proficient can you be operating like that. Both AirBus crashes were pilot error. You have to go back to the training they received and the manual flying abilities of the pilots flying.
 
I agree with you about the pilot error findings, Ron. But no airplane tail should fall apart because a pilot fully deflected the rudder pedals.

One of the maneuvers we consistently performed on my job was called a vertical polarization check. This was performed with a "dutch roll" dutch roll right and left 30 degrees to displace the horizontally mounted navigation antenna into the vertical plane, while at the same time continuing to fly the designated course precisely. This required full rudder with cross-controlled aileron at times to keep the aircraft on course. We did this in Saberliners, Jet Commanders, DC-3s, King Airs, Hawkers and Learjets to name a few types. We never had a vertical stabilizer fall off.

It's also common for small planes to perform a slips which sometimes require full rudder deflection. Their tails also remain on the airplane after the maneuver.
 
I agree with you about the pilot error findings, Ron. But no airplane tail should fall apart because a pilot fully deflected the rudder pedals......................

It's also common for small planes to perform slips which sometimes require full rudder deflection. Their tails also remain on the airplane after the maneuver.
Whew! I felt like the Lone Ranger for a minute...:blush:
 
I do agree with Boeing being a better aircraft. I know a few United pilots who call them selfs "BUNKIES" . Most of them really do not want to fly. Just do the minimum to keep their job. Many times they run out of landing currency and have to go get sim time to be legal.How proficient can you be operating like that. Both AirBus crashes were pilot error. You have to go back to the training they received and the manual flying abilities of the pilots flying.

This is how it goes. You fly 3 trips a month that have 2 landings each. The captain takes one...and the first officer takes one. If you are one of the relief pilots/IROs(I am assuming that is what "bunky" is referring to), then you get no landings. Occasionally, an FO or captain will give an IRO a landing. But you need 3 in the past 90 days....just like you do in ANY aircraft(Part 91...category and class). If you do not get those landings, then you go to the simulator to get them.

And it is not a matter of "wanting to fly" or "not wanting to fly". It is simply a matter of where you want to be.....in some hotel in Japan? Or with your loved ones on your boat? By doing long haul trips, you can knock out a full months worth of work in 9 days and have 21 days off. Now who wouldn't want that gig???
 
I agree with you about the pilot error findings, Ron. But no airplane tail should fall apart because a pilot fully deflected the rudder pedals.

Al, you have heard of the term "maneuvering speed"(Va), have you not??? I know you have...;) The whole premise of that speed is that below it, you can make full and abrupt control inputs without harm to the structure. Above that speed, then damage can occur. And transport category aircraft only have to make a full deflection IN ONE DIRECTION for the certification of the rudder at Va. On AA587, the FO made numerous full and abrupt inputs in opposing directions...well above Va.
 
I don't think we'll ever be changing ours. It's the reason we've stayed with conventional (cross-connected) yokes.


Two questions
  • Did the MD acquisition maintain this philosophy
  • Are the newer the Boeing designs sticking with conventional yokes
 
John, I agree that maneuvers involving full and abrupt control inputs above maneuvering speed can overstress components and lead to deformation or failure. It's an industry standard and should be well understood by all pilots. (I learned some new aspects of VA in reading the NTSB Accident Report.) But after this accident, Airbus issued a bulletin quoted below. The source for this is the NTSB Aircraft Accident Report
NTSB/AAR-04/04 PB2004-910404 Notation 7439B.


"1.18.4.1.2 Airbus Flight Crew Operating Manual Bulletin

In March 2002, Airbus issued A310/A300-600 FCOM Bulletin number 15/1,

“Subject No. 40, Use of Rudder on Transport Category Airplanes, in response to the Safety Board’s recommendations.” The bulletin emphasized proper operational use of the rudder and highlighted certification requirements and rudder control system design characteristics. The bulletin included the following information in a box labeled “CAUTION:”
Sudden commanded full, or nearly full, opposite rudder movement against a sideslip can generate loads that exceed the limit loads and possibly the ultimate loads and can result in structural failure. This is true even at speeds below the maximum design maneuvering speed, VA. (emphasis mine)

Certification regulations do not consider the loads imposed on the structure when there is sudden full, or nearly full, rudder movement that is opposite of the sideslip.

The bulletin also made the following operational recommendation:

RUDDERS SHOULD NOT BE USED:
– To induce roll, or
– To counter roll, induced by any type of turbulence.
Whatever the airborne flight condition may be, aggressive, full or nearly full, opposite rudder inputs must not be applied. Such inputs can lead to loads higher than the limit, or possibly the ultimate loads and can result in structural damage or failure.

The rudder travel limiter system is not designed to prevent structural damage or failure in the event of such rudder system inputs.

Note: Rudder reversals must never be incorporated into airline policy….[165]
As far as dutch roll is concerned, yaw damper action and natural aircraft damping are sufficient to adequately dampen dutch roll oscillations. The rudder should not be used to complement the yaw damper.

Note: Even if both yaw damper systems are lost, the rudders should not be used to dampen the dutch roll. Refer to the YAW DAMPER FAULT procedure.


At the public hearing on this accident, the American Airlines A300 fleet standards manager testified that American had not received “such specific limitations or prohibited maneuvers on the rudder use” before the flight 587 accident."

Here's a section discussing Pilot understanding on VA:

2.5.3 Pilot Guidance on Design Maneuvering Speed
During this accident investigation, the Safety Board learned that many pilots might have an incorrect understanding of the meaning of the design maneuvering speed (VA) and the extent of structural protection that exists when the airplane is operated below this speed.

From an engineering and design perspective, maneuvering speed is the maximum speed at which, from an initial 1 G flight condition, the airplane will be capable of sustaining an abrupt, full control input limited only by the stops or by maximum pilot effort. In designing airplanes to withstand these flight conditions, engineers consider each axis (pitch, roll, and yaw) individually and assume that, after a single full control input is made, the airplane is returned to stabilized flight conditions. Full inputs in more than one axis at the same time and multiple inputs in one axis are not considered in designing for these flight conditions.

The American Airlines managing director of flight operations technical told the Safety Board, during a postaccident interview, that most American Airlines pilots believed that the airplane would be protected from structural damage if alternating full rudder pedal inputs were made at an airspeed below maneuvering speed. The American Airlines A300
fleet standards manager confirmed this belief during testimony at the Board’s public hearing for this accident. The Board notes that the American Airlines A300 Operating Manual contained only one reference to design maneuvering speed, which indicated that it was the turbulence penetration speed (270 knots). However, as evidenced by flight 587, cyclic rudder pedal inputs, even when made at airspeeds below maneuvering speed, can result in catastrophic structural damage.
(emphasis mine)

My thoughts...

This accident report is a fascinating read and deals with many factors in the accident including inappropriate pilot rudder control inputs, light rudder force input/high sensitivity, rudder limiter design and limitations, vertical stabilizer attach methods and materials, pilot training, etc. Like most accidents, it's not just one failure that causes the accident, but a series of events/issues/failures that combine to cause the crash.

Thanks for posting this thread. Great stuff here.
 
At least Air Bust got thru the Paris Air Show with out crashing their new 350 .
 
Back
Top Bottom