AF447 wreckage found

[EMIT]

Contradiction?

xcitation #1973

No, xcitation, I was not contradicting myself. The discussion about manual flying, as is referred to in this thread, is about the so called danger from an aerodynamic standpoint, i.e. too great sensitivity of flight controls, or eeeeeek "coffin corner"!

The danger of inaccurate flying in RVSM is risk of collision. That is irrelevant in the discussion about AF447.

Are you not contradicting yourself - in one post you write about flying AT cruise altitude, in another one you query me about flying ABOVE cruise altitude.
To be sure, initially, the AF447 was AT cruise altitude, their real problem started when they went far above it through the setting of an unusually high pitch attitude.

Setting that unrealistic pitch attitude may be the result of never flying manual at high altitude, it is not the result of manual flying at altitude being somehow dangerous or exceedingly difficult.

And totempole, writing posts like this does not put me into the category of self aggrandized aces, I hope? Don't worry, my religion is to be highly proficient in automatics AND in manual.
Autopilots however, do not need training or recency to stay proficient - when they are one day unproficient, they announce it with a failure light. Pilots look physically the same when able and proficient or when not recent or unproficient. That's why I take every opportunity to fly manual and sure, some times or places are not opportune.

[chase888]

The challenge is to be totally proficient in correcting all the anomalies with the automatics entirely or partially. The supreme rule is never ever getting into situations where you have to use your super self declared " acy profiency " to manually get out of situations where you have gotten the automatics into. In my years of training pilots I have seen proud " old aces " get the automatics into ****ty situations and then proudly clicking them off to fly manually, mostly to an untidy end. Their pride.....they didn't crash despite the " failed automatics ". Actually there was absolutely nothing wrong with the automatics in those instances; the problem was the self declared aces did not understand the proper use of the automatics and misuse the functions.

I suspect Totempole has hit the 447 nail fairly on the head.
Not suggesting the crew did what is suggested in his post above, but its a fact that manual flying did not work out on 447 unfortunately.

[bubbers44]

The reason AF447 had the problem is when the AP threw the flying to the PF, he didn't respond properly and zoomed it up into a stall. Knowing how to hand fly at high altitude acurately would have saved the day but apparently he couldn't. I don't know how hard flying at altitude with a side stick is but it is quite doable with a yoke. Have done it quite a few times. Sometimes we dispatched with AP inop with FAA blessings.

[RWA]

quoting Bubbers44:-

"The reason AF447 had the problem is when the AP threw the flying to the PF, he didn't respond properly and zoomed it up into a stall. Knowing how to hand fly at high altitude acurately would have saved the day but apparently he couldn't."

Bubbers44, I don't think we can draw those sorts of conclusions from the fragmentary and incomplete information so far provided by the BEA. In fact, the more I read that Note the more convinced I become that it was deliberately framed to suggest 'pilot error,' without actually providing much evidence. However, a careful reading reveals that the PF did NOT in fact cause the 'zoom climb.'

The first relevant paragraph says:-

"From 2 h 10 min 05 , the autopilot then auto-thrust disengaged and the PF said "I have the controls". The airplane began to roll to the right and the PF made a left nose-up input. The stall warning sounded twice in a row. The recorded parameters show a sharp fall from about 275 kt to 60 kt in the speed displayed on the left primary flight display (PFD), then a few moments later in the speed displayed on the integrated standby instrument system (ISIS)."

So the first thing that happened was the autopilot and autothrust signing off, and the speeds (and very probably other instruments) going haywire. The PF made a left nose-up input, probably to counter the aeroplane's incipient roll and maintain altitude. Please note at this point that the time that movement was made was '2 h 10 min 05.'

The next paragraph of the note refers to a pitch-up some considerable time (at least eleven seconds) later. Please note especially the section I have 'bolded.'

"At 2 h 10 min 16, the PNF said "so, we’ve lost the speeds" then "alternate law […]". The airplane’s pitch attitude increased progressively beyond 10 degrees and the plane started to climb. The PF made nose-down control inputs and alternately left and right roll inputs. The vertical speed, which had reached 7,000 ft/min, dropped to 700 ft/min and the roll varied between 12 degrees right and 10 degrees left."

That paragraph does not refer to noseup inputs on the part of the PF before the 'zoom climb' - indeed, it says the complete opposite, that he countered the climb with nosedown inputs, and succeeded in more or less restoring level flight.

That, to my mind, leaves open the question of what caused the 'zoom climb' in the first place? The only thing one can say with any certainty is that, on the vestigial information provided by the BEA in the second quoted paragraph, it wasn't the pilot?

Then, of course, came the stall warning. And the PF seems to have responded by carrying out the prescribed drill at the time - 'TO/GA power and seek to maintain altitude.'

[goeasy]

The saddest realisation to me, on the limited info available, is that if this crew had resisted the urge to act, and just sat and watched, they may well all be alive today.

The hardest thing for any modern pilot today, is to NOT react to a screen full of warnings, and aural callouts, but to watch and fully assimilate the information available.

The 'hand-fly heros' may not have won this game either...

[cwatters]

The saddest realisation to me, on the limited info available, is that if this crew had resisted the urge to act, and just sat and watched, they may well all be alive today.

It's not clear that's the case.

[Ian W]

Zoom Climb to above coffin corner

if this crew had resisted the urge to act, and just sat and watched, they may well all be alive today

From reading all of this and associated threads, it seems that there have been 3 incidents reported on these threads of Airbus aircraft apparently undemanded 'zoom climbing'.

• The incident on approach to Paris may be different but resulted in the aircraft doing a wingover and recovering.
• The incident where an A340 climbed past an A330 in a zoom climb during the day VMC and recovered.
• AF447 appears to have zoom climbed to above the aerodynamic ceiling of the airframe. I am not sure that the system is designed to cope with suddenly being in the stall without any approach to it.

So it appears that something in the Airbus or handling can zoom climb the aircraft (has this happened with other types?)

I have only 'fallen out' above the aircraft ceiling in single jets where the aircraft were built for unusual positions and engine thrust had no pitch effect. If as everyone says the TOGA power would pitch the aircraft nose up, once that had been done and the aircraft started dropping stabilized but with an AOA that was on the wrong side of the drag curve, is it certain that there is sufficient aerodynamic authority in the controls to do a simple ND to recover?
Would the crew have to be more imaginative with large rudder deflection or even perhaps asymmetric thrust to put one wing down - and if they did that could they end up changing a stabilized stall into a spin?

It just seems to me that perhaps there was only one chance to recover from what happened and that was at the top of the zoom holding full nose down and reducing power which should have bunted the aircraft back into a flyable state. But the 'standard' stall recovery of NU to 5deg and TOGA was precisely the opposite.

[Lonewolf_50]

Practice makes perfect and if the company policy is to stay on automatics from lift off to near touch down, then that only leaves the flight simulator for honing manual flying skills.
Where simulator time is dictated by costs and company policy requires full use of automatics in the simulator, this leaves only a few minutes of effective time per session for the pilot to practice manually.

There's a problem. Use of full automatics in the sim. Seems to me that sim sessions ought to emphasize training scenarios one can't do in the aircraft.

For example, there are many operators who regard simulator practice at unusual attitude recoveries in IMC as a total waste of time. Often the instructor hasn't got a clue how to instruct on these manoeuvres anyway. Which is one reason why there has been an increase in loss of control in IMC accidents in recent years.
Until authorities bite the bullet and insist more emphasis be placed on pure flying skills versus automatic pilot monitoring skills, the current trend towards loss of control in IMC will not reverse

How can such a fundamental flying skill set be a waste of time?
Is this a symptom of institutional worship of autopilot?

This scares the hell out of me.

I am required to fly commercially next month to the east coast for a business meeting. You are suggesting that in some companies, skills necessary for IFR competency are being institutionally neglected, and allowed to atrophy. The flying public need to be warned.

RWA,
Then, of course, came the stall warning. And the PF seems to have responded by carrying out the prescribed drill at the time - 'TO/GA power and seek to maintain altitude.'
Ian W
It just seems to me that perhaps there was only one chance to recover from what happened and that was at the top of the zoom holding full nose down and reducing power which should have bunted the aircraft back into a flyable state. But the 'standard' stall recovery of NU to 5deg and TOGA was precisely the opposite

.
The training issue.
IF stall training is restricted to "near the ground environments" THEN the one time you stall at altitude you are playing catch up.

Based on a number of incidents related in the past two years on this topic, some crews in the past have caught up, this crew was unable to, as were some of the others in the past.

The question is, does BEA cover this in the final report? How much emhpasis?

We shall see.

[DC-ATE]

Lonewolf_50
I am required to fly commercially next month to the east coast for a business meeting.

There's always AMTRAK. Great way to travel even it it takes a little longer.

[Lonewolf_50]

I agree, but my company does not.

[cwatters]

How can such a fundamental flying skill set be a waste of time?

I agree with you but I wonder how many glider pilots would check the trim wheel position in the middle of recovering from a stall? Bet most would wait until they had recovered to S&L.

[bubbers44]

Autotrim on a sailplane?

[RWA]

Quoting Ian W:-

"The incident where an A340 climbed past an A330 in a zoom climb during the day VMC and recovered."

I think you've 'found the smoking gun,' Ian W, congratulations.

For 18 seconds after the autopilot disengaged the aircraft remained within 200 feet altitude of FL 360 but once AoA law was invoked at 14:21:50 hrs, the aircraft's attitude began to pitch nose-up. The pitch-up trend continued for 17 seconds reaching a peak of 15° nose-up shortly before the first nose-down sidestick command was applied. Throughout this phase the aircraft climbed rapidly (reaching a peak rate of about 6,000 ft/min) due to the increase in lift created by the flight control system's capture of alpha prot. The aircraft reached its apogee at FL 384 at 14:22:28 hrs where the airspeed had decayed to 205 KIAS and 0.67 Mach even though full thrust had been applied.

-----------

The crew subsequently descended back to FL 360 and successfully re-engaged the autopilot and autothrust systems.

That isn't referring to the AF 447 accident. It's from the British Air Accident Investigation Board's report on the 2000 A340 Atlantic 'zoom climb' incident you mention.

Apparently the autopilot and autothrust disengaged on that occasion too. The 'zoom climb' was triggered by the aircraft going into 'Angle of Attack' Law (yet another law for us to half-understand ).

http://www.rvs.uni-bielefeld.de/publ...pdf_501275.pdf

I find it very interesting that the 'protections' began to zoom-climb the A340 18 seconds after the autopilot and autothrust disengaged. That's just about exactly the same time that the AF 447 zoom climb commenced, as shown by the excerpts from the BEA Note that I quoted above.

[EMIT]

Read

On the quotes about the A-330 / A-340 zoom climb Airprox incident in NAT RVSM airspace:

Read the report, then you will find great differences with the AF447 case. This Airprox case was a situation with moderate to severe turbulence, large wind speed fluctuations that caused large speed (and Mach) variations and those initiated the incident. One such speed fluctuation caused AOA to increase to Alpha protect, etcetera.

There is no occurrence of the ERRONEOUS speed values that AF447 experienced, no LOSS of valid speeds, etcetra.

The cases are completely different, although the quotes try to make them similar.

[RWA]

Quoting EMIT:-

One such speed fluctuation caused AOA to increase to Alpha protect, etcetera.

There is no occurrence of the ERRONEOUS speed values that AF447 experienced, no LOSS of valid speeds, etcetra.

Oh dear, EMIT .

AF 447's speed readings were 'fluctuating' too; I don't somehow feel that it matters whether they were also 'erroneous' or not.

"The cases are completely different."

My own feeling is that there appear to have been only two obvious differences beween the two cases:-

1. The A340 only zoom-climbed at 6,000ft./min., whereas AF 447 zoom-climbed at 7,000.

2. The A340 crew were fortunate in that they were operating in daylight and no doubt had a visible horizon to work with.

But OK - if the 'systems' didn't cause the AF447 upset, what (in your opinion) did?

[before landing check list]

For 18 seconds after the autopilot disengaged the aircraft remained within 200 feet altitude of FL 360 but once AoA law was invoked at 14:21:50 hrs, the aircraft's attitude began to pitch nose-up. The pitch-up trend continued for 17 seconds reaching a peak of 15° nose-up shortly before the first nose-down sidestick command was applied. Throughout this phase the aircraft climbed rapidly (reaching a peak rate of about 6,000 ft/min) due to the increase in lift created by the flight control system's capture of alpha prot. The aircraft reached its apogee at FL 384 at 14:22:28 hrs where the airspeed had decayed to 205 KIAS and 0.67 Mach even though full thrust had been applied.

-----------
The crew subsequently descended back to FL 360 and successfully re-engaged the autopilot and autothrust systems

To tell you the truth I have no idea what "Alpha Prot" is nor does it really matter. I do know that for whatever reasons autopilots do strange things; due to a software error, stray trons floating around, faulty sensors etc. My question is this; Did the crew have access to reliable information as to what the aircraft was doing at the time either through the primary instrumentation or the standby and if so did they have control authority to put the aircraft where it needed to be or was it really out of control?
If they did have access to the actually flight path information of the aircraft and if they had access to the control of the aircraft why did they let it diverge so much for so long?
Now I am assuming they were night, IMC and kinetics should not used but if they knew something was amiss for whatever reason why did they not respond earlier? 18 seconds and close to 3000 vertical feet is a long time/way.

1. The A340 only zoom-climbed at 6,000ft./min., whereas AF 447 zoom/climbed at 7,000.

2. The A340 crew were fortunate in that they were operating in daylight and no doubt had a visible horizon to work with.

But OK - if the 'systems' didn't cause the AF447 upset, what (in your opinion) did?

If the systems did cause the initial upset but the pilots failed to react in a timely manner and control and correct flight path information was available to them, then poor training did them in. If the systems did cause the initial upset but control/flight path information was not available to them then it was/is a design flaw.

[cwatters]

Autotrim on a sailplane?

Thats the point, there isn't one so you don't get to learn how dangerous having one can be.

[Lonewolf_50]

RWA, from the discussions over at tech log, Alpha Protect works in normal law. At some point early in the AF 447 event, the pilots reported alternate law. At that point, Alpha Prot should not have been a factor in the behavior of the aircraft.

[EMIT]

IMC

Quote RWA

The A340 crew were fortunate in that they were operating in daylight and no doubt had a visible horizon to work with.

Remark EMIT: the Artificial Horizon is a perfect replacement for the Natural Horizon, the AI is used 100 % of the time during night and IMC conditions. A pitch of 15 degrees nose up is visibly crystal clear on an AI and is highly unusual in a civvie airliner at FL 350. The suggestion of another poster that maybe, the line Pitch 15, Thrust TOGA was inappropriately "quoted" inside the PF's head might hold some value. (the quoted pitch/power combination is the one to use in case of speed indication loss just after lift-off).

Quote RWA
But OK - if the 'systems' didn't cause the AF447 upset, what (in your opinion) did?

Remark EMIT: IF after all, the PF did not himself STEER the a/c into the zoom climb, then at least he should have seen the pitch change to 15 degr nose up, a very inappropriate pitch at that altitude, and should have tried to steer that pitch down to a normal value of 2 or 3 degrees above the horizon. This was not done, a lot of backstick was maintained.

Even though stall warnings, speeds or whatever were rightly or wrongly blaring, every big jet pilot must know that 15 degrees nose up at FL 350 is not gonna work in the long run (longer than a few seconds).

[bubbers44]

EMIT, hopefully the BEA final report will be a lot more informative than their last one. It lacked any meaningful information for some reason. They had all the data but it was like the guy who wrote it was getting off shift in 15 minutes and had a hot date and couldn't be bothered by details.