Air France 447 - On topic only!

[BoeingBobby]

O.K. Evan I just got off the phone with my friend that is an instructor at Airbus here in Miami for the last 16 years. I now have an understanding of what you have been trying to say. Gene explained the difference of NORMAL, ALTERNATE and DIRECT law to me and confirmed that as YOU said, there is a way to fly this aircraft not much differently than in the Boeing. He also told me they have tried to reproduce the situation in the simulator, but that the program does not simulate the high altitude regime very well. His opinion and that of his colleagues is that if the pilots had just pushed the nose down with the control stick it would have flown out of the stall. Good thing that comes of this is that he invited me to come down and fly the sim and get a feeling for the bus.

[Theoddkiwi]

Time out...Are we not talking indicated airspeed of 34 knots? I do believe that their airspeed was a good bit more, just not parallel to the earth nor the fuselage.

I'm betting that the AOA vane was working quite well in the fast slipstream of a fast "flat pancake" descent, even though we're only moving forward at 34 kts or only shoving 34 kts worth of pressure into the pitot tubes.

The main point of considering that the AOA is invalid at low indicated airspeeds is because they are considered to inaccurate.

I agree that they were probably physically and perfectly aligned with the 40deg AOA and the FDR does show that.

When Airbus designed this aircraft i doubt very much they thought it possible that a pilot could or would let their aircraft into such an outrageous attitutde. Let alone be in the air with an airspeed of 34kts.

As i said it just needs a minor change to add air/ground logic to the warning system and an appropriate Warning message.

[Gabriel]

I was thinking today about the plane's captain, and how he was at a significant disadvantage in terms of understanding what was going on.

He was absent when the autopilot went off and the PF pulled up. He didn't witness the first stall alarm go off, and then see the PF pull up again. Then he missed the critical 20 seconds, when the plane went from a climb to a steep descent, in essence, when it went from flying to not flying.

Add to that that the stall warning stopped shortly after the captain entered the cockpit and then it started to run erratically (the stall warning, no the captain).

As it was, he entered the cockpit when the aircraft was already stalled. He had no idea of what to look for in a stall, as to the best of my knowledge "stall recognition" is not taught, nor of course would he know what a stall "feels" or "sounds" like. While we may say that he still should have been able to put together the fact that the plane was falling precipitously along with the other stall symptoms, it certainly would not have been as straightforward a task as it would have been had he been present when the plane was put into the stall in the first place.

I can't accept that. I mean, maybe that's what happened, but it's not acceptable.

Any pilot at any level MUST know that 16º nose-up + 10,000 fpm down MEANS a stall, a very very big and deep one. Otherwise that pilot should read "Stick and Rudder" before further flight (a book that has some 60 years).

[Gabriel]

The main point of considering that the AOA is invalid at low indicated airspeeds is because they are considered to inaccurate.

I agree that they were probably physically and perfectly aligned with the 40deg AOA and the FDR does show that.

When Airbus designed this aircraft i doubt very much they thought it possible that a pilot could or would let their aircraft into such an outrageous attitutde. Let alone be in the air with an airspeed of 34kts.

As i said it just needs a minor change to add air/ground logic to the warning system and an appropriate Warning message.

That's because IT'S IMPOSSIBLE TO BE IN THE AIR WITH 34 KTS (or 60 for the matter). Not for more that 1 or 2 seconds while you change from going vertically straight up to vertically straight down (wherever the nose is ponting).

The only way to get below 60kts of measured airspeed with the plane in the air is with a WRONG AIRSPEED MEASUREMENT, like in this case.

Since Airbus made a UAS procedure, the UAS case was not unforeseen, and then to get a useless stall warning system when you need it most (in a UAS event, where you've got no speed indication and where you've lost all slow speed and stall protections but still remain with a control law that lacks AoA or speed stability) SHOULD HAVE BEEN AVOIDED PERIOD.

And since the only way to be travelling at REALLY 60 kts or less is to be in the ground, they could and should have used other means to inhibit it, like the air/ground logic based on weight-on-wheels, wheel spining, and radio altitude. It was really easy.

Not that this had to do a lot with this accident, since the crew had one full minute of healthy and uninterruped stall warning and still did nothing about it.

[Spectator]

AoA

The main point of considering that the AOA is invalid at low indicated airspeeds is because they are considered to inaccurate.

I agree that they were probably physically and perfectly aligned with the 40deg AOA and the FDR does show that.

When Airbus designed this aircraft i doubt very much they thought it possible that a pilot could or would let their aircraft into such an outrageous attitutde. Let alone be in the air with an airspeed of 34kts.

As i said it just needs a minor change to add air/ground logic to the warning system and an appropriate Warning message.

If AoA vanes are ever inaccurate do to lack of airspeed they should, if anything, be hanging down due to gravity and the inaccurate indication would be as if the 'plane was climbing something crazy. If, as in our case, the vanes are 40deg up, well... that wasn't gravity that did it now was it. Perhaps logic regarding positive/negative AoA readings could be added to the air/ground equation. Like if the vanes are 40deg up, you are either falling out the sky, or you are falling out the sky. In fact, I would expect that the stall warning would always only be triggered with the vanes up (as opposed to down) - perhaps its logic could be processed before the air/ground/<60kts invalidation filter?

[Theoddkiwi]

If AoA vanes are ever inaccurate do to lack of airspeed they should, if anything, be hanging down due to gravity and the inaccurate indication would be as if the 'plane was climbing something crazy. If, as in our case, the vanes are 40deg up, well... that wasn't gravity that did it now was it. Perhaps logic regarding positive/negative AoA readings could be added to the air/ground equation. Like if the vanes are 40deg up, you are either falling out the sky, or you are falling out the sky. In fact, I would expect that the stall warning would always only be triggered with the vanes up (as opposed to down) - perhaps its logic could be processed before the air/ground/<60kts invalidation filter?

No the vanes don't drop down when on the ground or at low speeds, they are finely balanced. You can grab a vane and move it to any position and it will stay there. They have a freedom of movement of +/-90 degrees. The vanes for our aircraft end up pointing in all sorts of directions after landing, probably due to the aircraft taxiing over bumps etc.

This might also be the inaccuracy that occurs at less the 60kts, perhaps G effects can cause them to be in accurate.

[Theoddkiwi]

That's because IT'S IMPOSSIBLE TO BE IN THE AIR WITH 34 KTS (or 60 for the matter). Not for more that 1 or 2 seconds while you change from going vertically straight up to vertically straight down (wherever the nose is ponting).

The only way to get below 60kts of measured airspeed with the plane in the air is with a WRONG AIRSPEED MEASUREMENT, like in this case.

Since Airbus made a UAS procedure, the UAS case was not unforeseen, and then to get a useless stall warning system when you need it most (in a UAS event, where you've got no speed indication and where you've lost all slow speed and stall protections but still remain with a control law that lacks AoA or speed stability) SHOULD HAVE BEEN AVOIDED PERIOD.

And since the only way to be travelling at REALLY 60 kts or less is to be in the ground, they could and should have used other means to inhibit it, like the air/ground logic based on weight-on-wheels, wheel spining, and radio altitude. It was really easy.

Not that this had to do a lot with this accident, since the crew had one full minute of healthy and uninterruped stall warning and still did nothing about it.

I agree it should be impossible to have an airspeed of less than 60kts whilst in the air. I guess in this case 34kts was the pitot pressure acting on the probes in a forward direction, but due to a 40degree attitude the actual speed through the air in a vertical attitude must have been much higher as i would think Pitot tubes don't really measure the pressure to well if its coming from the side or below in this case.

I agree that Airbus did anticipate UAS and had a clear and simple procadure to deal with it and to me is the primary aspect of this crash.

[Spectator]

No the vanes don't drop down when on the ground or at low speeds, they are finely balanced. You can grab a vane and move it to any position and it will stay there. They have a freedom of movement of +/-90 degrees. The vanes for our aircraft end up pointing in all sorts of directions after landing, probably due to the aircraft taxiing over bumps etc.

Thanks, that puts paid to that idea.

[3WE]

...His opinion and that of his colleagues is that if the pilots had just pushed the nose down with the control stick it would have flown out of the stall...

They are all asshats and idiots.

[Evan]

Since Airbus made a UAS procedure, the UAS case was not unforeseen, and then to get a useless stall warning system when you need it most (in a UAS event, where you've got no speed indication and where you've lost all slow speed and stall protections but still remain with a control law that lacks AoA or speed stability) SHOULD HAVE BEEN AVOIDED PERIOD.

And since the only way to be travelling at REALLY 60 kts or less is to be in the ground, they could and should have used other means to inhibit it, like the air/ground logic based on weight-on-wheels, wheel spining, and radio altitude. It was really easy.

The primary function of the AoA sensors is NOT to inform the pilots or provide cockpit instrumentation (at least not yet) but to provide data to autoflight systems and computer protections. When this data becomes "invalid", this means that these systems can no longer consider it reliable. This is for VERY GOOD REASONS. Since the Qantas ADIRU incident, we all know what can happen if the wrong AoA is considered valid. Since the Colgan incident, we all know what can happen when false stall warnings occur.

The aircraft has no way of knowing why the speeds are indicating below 60kts, but it does know that the AoA vanes are not accurate below that speed. So it rules them out. For safety reasons, the Airbus philosophy that allows for FBW authority REQUIRES that any air or attitude data that is not considered reliable MUST be ruled out. When the computers are not certain of the incoming data, they hand the aircraft over to the pilots. Perhaps unreliable AoA should still drive the stall warning, but then see: COLGAN 3407.

As we have discussed an nauseam, the pilots do not need AoA or stall warnings to avoid or recover from a stall. When airspeed data is unreliable and may be producing false indications below 60kts, the procedure is clearly outlined. This procedure CANNOT result in a stall AoA. Airspeed disagree typically lasts only a minute or two. When you WEIGH all of these factors, I can understand why Airbus made the design decisions that it made. Contrary to popular myth, Airbus engineers DID assume pilot competency when designing the aircraft.

I see your point however. If we are going to put AoA indications back in the cockpit, we are going to have to have at least one sensor that sends data to that instrument at all times (but not to the systems!) and perhaps provides data for the stall announcement, maybe something like "CHECK FOR STALL" when the AoA is considered unreliable, and a message next to the indicator that flashes AOA UNRELIABLE.

But such an alert should also be triggered by a combination of high attitude and sink rate and/or a high rate of acceleration (IR data). But of course, that alert should already be triggered in the pilot's mind...

You have to admit, it's a difficult design decision, as each option has an inherent danger. I think Airbus knows that you can't idiot-proof a plane. I think any change here requires a great deal of consideration.

Meanwhile, the AF pilot's union uses this as a weak excuse to avoid the glaringly obvious reality. They seem to have no shame.

[dkmax]

Zero Airspeed

Hi Guys,

I've followed this thread from the start, as fascinated as everone to try and understand how an A330 can just fall out of the sky!? The way the evidence is now pointing is quite staggering!?

I have a laymen question if I may...

With airspeed as low as 60 or even 34 knots, I just wondered if the plane had passed the point of no return? Do the control surfaces have any effect at such a low speed?

Hypothetically, if you could hoist an A330 up to 38,000ft, 0 airspeed, wings level, minimal AoA, TOGA & simply 'drop it'...

Would it fly? I guess not, but could you actually get the nose down to make it fly?

[BoeingBobby]

hi guys,

i've followed this thread from the start, as fascinated as everone to try and understand how an a330 can just fall out of the sky!? The way the evidence is now pointing is quite staggering!?

I have a laymen question if i may...

With airspeed as low as 60 or even 34 knots, i just wondered if the plane had passed the point of no return? Do the control surfaces have any effect at such a low speed?

Hypothetically, if you could hoist an a330 up to 38,000ft, 0 airspeed, wings level, minimal aoa, toga & simply 'drop it'...

Would it fly? I guess not, but could you actually get the nose down to make it fly?

yes!

[BoeingBobby]

They are all asshats and idiots.

Obviously you have just left the hospital after doing brain surgery for the last 6 hours and you are a bit tired Doctor!

[mawheatley]

Does anyone know where the crew rest area is/was on this aircraft?

About 10yrs ago I was a F/A for a US carrier and we flew A330's. I only got two trips on the A330, preferring the 767 myself. (Well the 767 went to London - the company was paying me to go home twice a week! ) On both of those types however the crew rest was one of the front rows in Business Class. Now our A330's were 3 class configuration, meaning that to get from crew rest to the flight deck the pilots had to walk through some of Business Class and ALL of First Class.

That being said, it may have no bearing on how Air France A330's are configured, but either way, he had to walk forward, therefore (presumably) uphill?

[Evan]

About 10yrs ago I was a F/A for a US carrier and we flew A330's. I only got two trips on the A330, preferring the 767 myself. (Well the 767 went to London - the company was paying me to go home twice a week! ) On both of those types however the crew rest was one of the front rows in Business Class. Now our A330's were 3 class configuration, meaning that to get from crew rest to the flight deck the pilots had to walk through some of Business Class and ALL of First Class.

That being said, it may have no bearing on how Air France A330's are configured, but either way, he had to walk forward, therefore (presumably) uphill?

It's in a modified cargo compartment on the cargo deck, presumably forward.