Air France Flight 447 - finally solved.

[Hellis 0]

Quote

>How can one become a pilot of such an airplane and not know what a stall is and how
>to fix it?

Because it's a plane that, normally, cannot stall. When you spend much of your career flying an aircraft that can't stall, you're going to tend to lose your ability to predict, detect and deal with a stall.

I understand what you say but...

Quote

even though the word "Stall!" will blare through the cockpit 75 times.

That would not be a good clue whats going on?
Even if the airplane *should* not be able to stall, the pilots should atleast have commented the fact that the airplane keept telling them it's stalling.

Just ignoring it because the computer won't allow it to stall sounds like a very bad idea, and I guess it was.

[billvon 2,396]

>That would not be a good clue whats going on?

People have the ability to block out all extraneous information when they're in trouble and they think they need to concentrate on a few things. I've seen this hundreds of times in AFF. People stare at their altimeters, but don't read what it's saying, don't see hand signals and don't see the ground coming up. People stare at me and don't look at their altimeters. People stare at the ground and ignore the guy screaming at them over the radio.

And I have a feeling that if I were flying, started having problems and the computer started saying "Caler! Caler!" I might file that under "I'll figure out what it's saying after I deal with this problem."

>Just ignoring it because the computer won't allow it to stall sounds like a very bad
>idea, and I guess it was.

Yep.

Most larger aircraft have stick-shakers to indicate an incipient stall. Maybe it's time to add a stick shocker, one that delivers progressively more and more severe shocks to the pilot the longer he has the stick full back.

[nigel99 143]

Quote

Under "normal law" the computers will keep the airplane in its flight envelope no matter what the pilot does with the controls. No matter how hard the pilot pulls back on the stick, no matter how slow he tries to fly, the computers will simply override the pilot and keep the nose where it needs to be to maintain a safe AOA.

Thanks for your detailed reply and interesting info. I realise that the computer imposes the limits rather than the design of the aircraft aerodynamics.

Experienced jumper - someone who has made mistakes more often than I have and lived.

[Rookie120 0]

Quote

Maybe it's time to add a stick shocker, one that delivers progressively more and more severe shocks to the pilot the longer he has the stick full back.

Hell a lot of planes as you know have stick pushers now. I remember in the CRJ the amount of force to override the pusher would damn near pull me out of the seat and I weigh 185lbs. It's one hell of a tug.

If you find yourself in a fair fight, your tactics suck!

[piisfish 135]

Quote

Slow down too much put the nose too high and watch!

An-2 anyone ?

scissors beat paper, paper beat rock, rock beat wingsuit - KarlM

[shah269 0]

Quote

Quote

Maybe it's time to add a stick shocker, one that delivers progressively more and more severe shocks to the pilot the longer he has the stick full back.

Hell a lot of planes as you know have stick pushers now. I remember in the CRJ the amount of force to override the pusher would damn near pull me out of the seat and I weigh 185lbs. It's one hell of a tug.

I had to read the article twice...how sad!

Dumb question is this a failure of training or just too many odd things going wrong at the same time?

Either way

Life through good thoughts, good words, and good deeds is necessary to ensure happiness and to keep chaos at bay.

The only thing that falls from the sky is birdshit and fools!

[diablopilot 2]

Quote

>How can one become a pilot of such an airplane and not know what a stall is and how
>to fix it?

Because it's a plane that, normally, cannot stall. When you spend much of your career flying an aircraft that can't stall, you're going to tend to lose your ability to predict, detect and deal with a stall.

And right there is your argument made against these advanced systems. When you take the pilot out of the loop bad things happen.

----------------------------------------------
You're not as good as you think you are. Seriously.

[piisfish 135]

Quote

And right there is your argument made against these advanced systems. When you take the pilot out of the loop bad things happen.

the problem is you are talking about PILOT , and these people are now System Operators/Supervsiors. Not Pilots/Flyers. I know a couple of these Air-Bus drivers who actually do not enjoy flying a plane... It is just a job. They could as well be driving a train/truck/dumpster if it paid the same....

scissors beat paper, paper beat rock, rock beat wingsuit - KarlM

[Calvin19 0]

Quote

Quote

And right there is your argument made against these advanced systems. When you take the pilot out of the loop bad things happen.

the problem is you are talking about PILOT , and these people are now System Operators/Supervsiors. Not Pilots/Flyers. I know a couple of these Air-Bus drivers who actually do not enjoy flying a plane... It is just a job. They could as well be driving a train/truck/dumpster if it paid the same....

YES!

Did you guys read the transcript? It's AWESOME.



"we totally lost control!" "We've tried everything!"
"Damn it, we're going to crash... This can't be happening!"
"but what's happening?!"

[shropshire 0]

Quote

Quote

Quote

And right there is your argument made against these advanced systems. When you take the pilot out of the loop bad things happen.

the problem is you are talking about PILOT , and these people are now System Operators/Supervsiors. Not Pilots/Flyers. I know a couple of these Air-Bus drivers who actually do not enjoy flying a plane... It is just a job. They could as well be driving a train/truck/dumpster if it paid the same....

YES!

Did you guys read the transcript? It's AWESOME.



"we totally lost control!" "We've tried everything!"
"Damn it, we're going to crash... This can't be happening!"
"but what's happening?!"

Sounds weird but I had to read it a few times and still wanted the outcome to be different.

(.)Y(.)
Chivalry is not dead; it only sleeps for want of work to do. - Jerome K Jerome

[sundevil777 93]

Quote

>How can one become a pilot of such an airplane and not know what a stall is and how
>to fix it?

Because it's a plane that, normally, cannot stall. When you spend much of your career flying an aircraft that can't stall, you're going to tend to lose your ability to predict, detect and deal with a stall.

Normally is the operative word. Even the Airbus test pilots have been wrong about how well the A330 could prevent a stall.

Airbus was so confident of their 'envelope protection' that it caused a crash of an A330 many years ago. An airline CEO and top pilot were being treated to a demonstration flight. Airbus decided to impress them with how the plane could not be made to stall, so they planned to take off very aggressively and immediately idle one engine and turn off one of the hydraulic systems. However the flight control engineers had never figured on such an aggressive, very high pitch rate climb being combined with the other 'failures', and the plane could not stop the stall. They ran out of altitude. The investigation found fault in how the Airbus pilots decided to create their own scenario during the pre flight briefing that would impress the customer, completely confident that the plane could not be stalled (as they had been claiming). This incident received much less attention than the crash of the A320 at the Paris air show because no normal passengers were on board.

I have a few issues with the Airbus cockpit (this applies to all of their fly-by-wire planes A320/A330/A340/A380 series also, things that I think are a bad idea for a commercial airliner):

1) The sidestick controllers

They give no feedback when the autopilot is flying - no movement. I think that seeing the control column/wheel move is a great way for the pilot to know what the autopilot is trying to do.

There is also no force feedback as you try to make the plane do more severe maneuvers. I'm not absolutely sure if there is a stick shaker, but I think not.

When a pilot moves from right to left seat, they have to 'train' their other arm/hand to use the sidestick controller (it is on the outboard side). I think this is inherently a bad idea, for a commercial airliner.

The sidestick controllers are not linked, so when one pilot is flying, the other stick does not move. So what happens if both are trying to fly? There is software logic that decides this, and it is indicated which side is in control with an indicator light, and the pilots can override/take command on their side with the push of a switch. The thought of the logic not working and the pilots pushing buttons to take command is not comforting to me.

2) The autothrottle

They do not move in the normal mode, meaning that as the autopilot adjusts power, the handles stay in one position. You can see by instruments what it is trying to do, but not by movement of the big handles. In the normal mode, it is just a big rotary switch. Again, I think seeing the throttles move is a great way for the pilot to know what the autopilot/autothrottle system is trying to do.

3) Envelope protection/limiting (fancy term for the computer doesn't let you do stupid maneuvers or overstress the plane)

I don't think it is a good idea to have the fly-by-wire system prevent the pilot from putting too much stress on the plane. We are talking about a airliner, not a fighter. The pilots will never overstress it just for practice. If they need to go beyond the design structural limits to try to get out of a dive, for instance, then I say they should be allowed to try, without having to take the time to put it in an alternate law mode.

The Boeing fly by wire system on the 777 still uses a traditional control column/wheel. There are some distinct advantages to the sidesticks, such as reduced weight, and much more room in front of the pilot for a worktable to use maps/manuals/etc, It was tempting, however Boeing decided that it was a bad tradeoff overall.

Boeing also decided that strict envelope protection was a bad idea, and that increasing force feedback (it is 'artificial') was a better way of preventing pilots from doing stupid things while not limiting their ability to fly the plane how they see fit. Both Boeing and Airbus have a switch on the overhead console that puts them into a 'direct' mode that cuts out the complex flight control logic and uses separate/simpler software that results in a less 'refined' feeling for the pilot. This requires time to activate it, and I would rather the pilots have complete control to try what they want without having to look up and back to activate a switch first.

If I've incorrectly described some details of how these systems worked, I apologize in advance.

It probably will not surprise you to learn that I used to be an engineer for Boeing in their flight deck group during the development of the 777, hence the 777 in my username.

Of couse Boeing has had problems, such as 737 rudder reversals, 747 cargo door latches, and uncommanded 767 thrust reversals. However, I think their choice on the column over the sidestick controller, and the flight control logic are examples of interesting, and important points of discussion.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[sundevil777 93]

To reinforce my comment that side stick controllers aren't the best for an airliner, from the article's transcript:

Quote

Unlike the control yokes of a Boeing jetliner, the side sticks on an Airbus are "asynchronous"—that is, they move independently. "If the person in the right seat is pulling back on the joystick, the person in the left seat doesn't feel it," says Dr. David Esser, a professor of aeronautical science at Embry-Riddle Aeronautical University. "Their stick doesn't move just because the other one does, unlike the old-fashioned mechanical systems like you find in small planes, where if you turn one, the [other] one turns the same way." Robert has no idea that, despite their conversation about descending, Bonin has continued to pull back on the side stick.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[sundevil777 93]

Quote

Most larger aircraft have stick-shakers to indicate an incipient stall. Maybe it's time to add a stick shocker, one that delivers progressively more and more severe shocks to the pilot the longer he has the stick full back.

Only the larger aircraft that have a traditional wheel and column. All the thousands of modern Airbus airliners with full fly by wire systems do not have it. Since the planes have no provisions in the design to push the sidestick controller, adding a stick shaker would not be easy, and as can be seen from what's happened so far, is not being considered. Maybe an electrical shock to the seat would be more easily implemented.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[Rookie120 0]

Quote

Dumb question is this a failure of training or just too many odd things going wrong at the same time?

I dont know. The final report is not out yet. That probably will take a couple years. But from most crashes I have read about or incidents it never seems to be 1 thing that caused a crash. It always seems to be a line of things that. It seems that with possible pitot or static ports iced up, flight computers going to alternate modes, Capt not in the seat, see it's a list of things that contribute but the final report is what I will read. All I am reading now is stuff that could have caused and led up to the crash.

If you find yourself in a fair fight, your tactics suck!

[billvon 2,396]

Agreed on most of what you said. Some comments:

>1) The sidestick controllers

I don't think the problem here is that they are sidestick controllers - the problems are (as you mentioned) lack of force feedback, lack of relationship between stick displacement and control surface action and lack of coupling between the two sticks. If you add those things back into the mix I don't think they are any worse than a traditional yoke. (Witness the success of the SR22.)

>3) Envelope protection/limiting (fancy term for the computer doesn't let you do
>stupid maneuvers or overstress the plane)

I think envelope protection does have an important function (it can prevent incidents like AA587) but as you mention there are good ways and bad ways to implement it. The problem with the Airbus method, I think, is summed up by an often-heard exclamation on CVR tapes after incidents both minor and major - "what's it doing now?"

[Deisel 35]

When this filght went down among the initial assumptions was one of a 5 minute nose dive from over 30k ft. That one stood out for me. At least now we know that it didnt happen that way. Hell, its even presumable that some of these poor souls were catching some shut-eye at 0200 and just never woke up. At least I hope so. BSBD to all on board

The brave may not live forever, but the timid never live at all.

[sundevil777 93]

Quote

I don't think the problem here is that they are sidestick controllers - the problems are (as you mentioned) lack of force feedback, lack of relationship between stick displacement and control surface action and lack of coupling between the two sticks. If you add those things back into the mix I don't think they are any worse than a traditional yoke. (Witness the success of the SR22.)

The SR22 has side sticks with feedback?

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[sacex250 0]

Quote

Quote

I don't think the problem here is that they are sidestick controllers - the problems are (as you mentioned) lack of force feedback, lack of relationship between stick displacement and control surface action and lack of coupling between the two sticks. If you add those things back into the mix I don't think they are any worse than a traditional yoke. (Witness the success of the SR22.)

The SR22 has side sticks with feedback?

Yes, an SR22 has feedback in the control sticks. The farther you move the controls the harder it pushes back on you because of the aerodynamic forces on the control surfaces.

In an Airbus, the pilot can't feel what the controls are doing. The pilot could have the stick pushed all the way to the right but the FBW could have the ailerons in any other postiion and the pilot wouldn't know it. The FBW system doesn't tell the pilots what it's doing.

In AF447, the PF had his side stick full back but the other pilot had no way of knowing it.

It's all been said before, no sense repeating it here.

[Calvin19 0]

Quote

Quote

I don't think the problem here is that they are sidestick controllers - the problems are (as you mentioned) lack of force feedback, lack of relationship between stick displacement and control surface action and lack of coupling between the two sticks. If you add those things back into the mix I don't think they are any worse than a traditional yoke. (Witness the success of the SR22.)

The SR22 has side sticks with feedback?

The Cirrus is a 3000 pound general aviation airplane, the side sticks are hard-coupled to the surfaces, just like pretty much any jump aircraft anyone here jumps from. It also has heated seats and a magic reserve parachute for safe intentional continued VFR flight into IFR conditions.

The A330 is a half million pound (500,000lb) transport aircraft.

[sundevil777 93]

Quote

Quote

Quote

I don't think the problem here is that they are sidestick controllers - the problems are (as you mentioned) lack of force feedback, lack of relationship between stick displacement and control surface action and lack of coupling between the two sticks. If you add those things back into the mix I don't think they are any worse than a traditional yoke. (Witness the success of the SR22.)

The SR22 has side sticks with feedback?

The Cirrus is a 3000 pound general aviation airplane, the side sticks are hard-coupled to the surfaces, just like pretty much any jump aircraft anyone here jumps from. It also has heated seats and a magic reserve parachute for safe intentional continued VFR flight into IFR conditions.

The A330 is a half million pound (500,000lb) transport aircraft.

So, the SR22 is a side stick, but is not fly by wire - it is hooked up by cables, etc. to the control surfaces? That is a crappy analogy to the point of how the A330 flight deck works (Billvon's crappy analogy). So to put it more precisely, Fly by wire airliners are not so well served by side stick controllers.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[sundevil777 93]

Quote

In an Airbus, the pilot can't feel what the controls are doing. The pilot could have the stick pushed all the way to the right but the FBW could have the ailerons in any other postiion and the pilot wouldn't know it. The FBW system doesn't tell the pilots what it's doing.

In AF447, the PF had his side stick full back but the other pilot had no way of knowing it.

Yes, as I had mentioned in my post. The Airbus designers took the established path for their fly by wire design with side stick controllers - no feedback and no coupling between the two sides (those two decisions come along easily together). That philosophy was also applied to the auto throttle.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[Calvin19 0]

I don't think Billvon's description was bad at all. He accurately described the fly by wire side stick system and it's drawbacks in that airplane.


He also made an accurate comparison of the overly-modernized aircraft the A340 and the SR20/22. Both airplanes give their pilots a false sense of security and illusion of automation.

-SPACE-

[sundevil777 93]

Quote

I don't think Billvon's description was bad at all. He accurately described the fly by wire side stick system and it's drawbacks in that airplane.


He also made an accurate comparison of the overly-modernized aircraft the A340 and the SR20/22. Both airplanes give their pilots a false sense of security and illusion of automation.

I like to pick on Bill for crappy analogies.

I described the system's drawbacks, he mentioned that the SR has sidestick controllers that give feedback. Because the SR is not fly by wire, the comparison is not useful.

I don't see how the SR would give a pilot a false sense of security and illusion of automation. Perhaps you're thinking it is fly by wire. I don't think it is.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[Calvin19 0]

Quote

I don't see how the SR would give a pilot a false sense of security and illusion of automation. Perhaps you're thinking it is fly by wire. I don't think it is.

False sense of security, absolutely. Maybe not an especially obvious illusion of automation. But in reality it is only a normal airplane with a very modernized avionics system. It is susceptible to all the same real world dangers of navigation and aerodynamics that any other airplane is. This includes flying into a mountain goat, over-speeding and stall/spinning, all of which seem to occur a disproportional amount of Cirrus pilots.


http://www.aopa.org/asf/asfarticles/2004/sp0402.html

http://news.injuryboard.com/cirrus-fatalities-have-critics-questioning-safety.aspx?googleid=262482

[Calvin19 0]

Quote

I think envelope protection does have an important function (it can prevent incidents like AA587) but as you mention there are good ways and bad ways to implement it.

Remember, 587 was also arguably caused by grossly insignificant or non-existant control feedback. In both 587 and 447 it could be assumed that control feedback would have easily prevented the pilots from making the mistakes they did.

"Control forces should not be so high that the pilot cannot safely maneuver the airplane. Also the forces should not be so light that it would take exceptional skill to maneuver the airplane without overstressing it or losing control. The airplane response to any control input should be predictable to t he pilot.”
FAA Advisory Circular 25-7A