Hi,

Another graphic chronology events with ACARS included:

Click link
imgur: the simple image sharer

I've suggested the same. Ok, flying into a weather front - as they did - could mean rapid changes in both direction and speed of winds. Yet having an advisory reading on your display, perhaps showing GS, IAS and their difference side by side in colored numbers could be, I'm imaging, a handy tool when things start to go wrong.

Why the backstick?
 

The consensus seems to be an incredulity as to why on Earth the PF held backstick? None of us would ever do that.

Speculation: PF felt the stall warning was triggered by a 10000'/min downdraft -- and just before, he felt he had hit a 7000'/min updraft that he had countered with forward stick. Perhaps PF hung his hat on his windshear/terrain avoidance training.

When in a rapid descent, a low airspeed indicates stall which requires stick forward while increasing airspeed indicates spiral dive which requires wings level and stick back (leaving out the finer details of spiral dive recoveries for sake of brevity).

But here the crew does not have an airspeed worth looking at as it is flopping back and forth.

Recovering from an upset at night in IMC is demanding enough. When you don't have a believable airspeed to distinguish between stall and spiral dive, you have a 50/50 chance of getting it wrong.

In normal weather in a spamcan or glider, you can hear increasing airspeed. Not so easy in a CB.

chron
 

jcjeant
do you have the spreadsheet, or just the image?

Devionics:
 

"listen to what ArnieG has to say..."
I got that talk in print. To a large part rubbish. Because it is in my own area of specaialist competence it was sad to read. Many statements made are wrong, e.g. that PWR:s dont't have containment venting systems, In Sweden I know of three PWR:s which have venting systems connected to venturi scrubbers.
The problems at Fukoshima seems to have been that their "hydrogen recombiners" needed electrical supply to operate. Electricity as assumed to be available from four sources - batteries, diesel generators, perhaps house turbine operation, and perhaps even the main grid. All had to fail during the same time.
I sincerely hope that most of you have better knowledge in the field of aircrafts and flying.

Regards

There's an aspect of the BEA report no-one has yet commented on on this thread.

It states: "The PF made an input on the sidestick to the left and nose-up stops, which lasted about 30 seconds." [my emphasis]

My questions are:

1. What would be the effect of prolonged application of full roll deflection in a stalled A330, and
2. What might lead the PF to make such input?

Hi,
This is a usefull job except for the time stamping of ACARS.
Those messages should not be based on such a precise reception time stamp as it is different from when it was triggered (and in what particular order), but rather by its time stamping from its emission time which is the nearest minute (0210, 0211, 0212...). If displayed like that, it may conduct to a very bad interpretation in relation with real time cockpit events.

Hi,

Of course .. but who know the exact realtime stamp in situ of the ACARS ?

Image only

RBF

Surely in a spiral dive you would have increasing G if you pulled backstick which you could sense. Can't see the confusion myself, certainly not for 4 min's.

It may well be when all is said and done that we will have to accept that the crew got it wrong, in failing to avoid the weather (if that is what it was) and the Captain in his bunk and in how they dealt with the failure's when they occurred. Not a pleasent thought but there it is. What we need to ask ourselves, each individually, is if it could happen to them could it happen to me. I would like to think that I could have dealt with it, then again I would like to think that 3 trained pilots flying for a major carrier could have dealt with it They didn't. Time for me and all of us to review a few things.

Ok, got it. But you just pulled the same stunt...and you just joined....let's move on...

EVERY single post on that site seems to be by the same, vitriolically anti-Airbus poster. Caveat emptor, I'd say.

Do the inboard weather radar can estimate air speed ?

backup?
 

It seems odd that loss of the pitots should cause the whole system to become unworkable. Is there any sort of backup, eg inertial? I know there are three pitots, but if one can be overwhelmed by ice, then all can, and so three are no better than one. And if the FDR knows the pitch attitude, the angle of attack, and the rate of descent, why don't the pilots? The airplane seems to have had plenty of forward speed: 10,000 fpm vertical = 99 knots, with a flight path angle of -25 degrees (40 aoa - 15 pitch) suggests a TAS of 99/sin(25) or 234 kt (is this right? -- somebody help me out here) and it was responsive to roll inputs. I completely understand and sympathize with pilot confusion and mental paralysis, having experienced it myself, but it seems as if there are big issues here of information presentation and human/system interface, notably the counterintuitive behavior of the stall warning. We know, and the FDR seems to have known, that the airplane was stalled; why wasn't there a nice firm female voice telling the crew so the whole time?

no, it cannot.

And i don't know if they had any GPS, but IAS and GS at that FL are so different that it would be useless.

You don't know the "exact" realtime, but the "nearest minute". Quite simply, when you don't know something (a precise factor), you don't make it up. In this case, you can only list those ACARS, in no particular order, by minute with what information the CMC provided (0210, 0211, 0212..) and assume the time range boundaries for each to be sent.

Hence, you can't have a direct relation with any particular "second" of the real time scale :
- if an ACARS was CMC stamped at 0210 and stamp recieved at 0210:26, all you know is that it could have been triggered between 0209:31 (the nearest minute) and 0210:20 (as you will allow about 6 seconds for processing). If the same was recieved at 0211:31, it would be meaningless to look outside the 0209:31-0210:30 window in order to understand what caused it (in this case, there is certainly no relation with events past 0210:31 even if it was recieved one minute later).

And isn't it interesting that these "vitriolically anti-Airbus posters" always manage to have 'sources' in the investigation that get 'leaks' that -just happen- damn Airbus... but later when the real facts come out, the 'leak' was poppycock.

It's almost as if they are sitting in their mom's basement making stuff up.... not that we'd ever see anyone do THAT on the internet. ;)

'At 1 h 55, the Captain woke the second co-pilot and said "… he’s going to take my place" Who was he referring to? The DailyTelegraph reports ....'according to flight recorder data, the younger of the two men, Pierre-Cédric Bonin, 32, angled the jet's nose higher'. David Robert, 37 was the other FO, and he was licensed to replace the captain. So if M. Bonin was nominated by captain Dubois to take over, then why as it appears he was not licensed to replace the captain, as per the BEA Interim report.

I did not read the previous hundreds of post, but it seems the crew had an entirely different perception of what was happening during the last minutes.

It seems if they had a better awareness they could "easily" have regained control, right? (pushing the stick)

I'm admit to be lazy here , but isn't there any speed measurement backup procedure if e.g. the pitots block/freeze? (e.g. GPS, Iphone ? ) It seems they had little time but some minutes..

The 330 in better days..
JetPhotos.Net Photo » F-GZCP (CN: 660) Air France Airbus A330-203 by Aurélien TRANCHET

This was exactly my post in 2nd thread. Although for actual flying you do need True Air Speed which is the speed of aircraft relative to given part of the atmosphere, but inertial navigation system should give a very good clue to onboard computer by telling it:

- Hey HAL your air speed dropped from 275knots to 60 knots WITHOUT any decceleration!

-So disregard your speed change and keep the current flying parameters - thrust and pitch. Until deceleration/acceleration tells it that speed changed.

And then give pilots info about unrelaible airspeed.

Apologies to anyone with a pilot’s licence or knowledge of aerodynamics.

To everybody else on this thread please listen;

Ground speed is only relevant to an aircraft when you come into contact with the ground (or want to know at what time you will arrive at your destination)

Air is a fluid medium. Whilst airborne air speed is the key to staying there.

When assessing student pilots grasp of this concept I used to ask them a simple question; The wind is 270 degrees at 20 kts and I release a helium filled toy balloon, where will it be in one hour?

During instrument training the mantra is “believe your instruments but keep the scan going” in other words; don’t fixate on one, the primary instruments will tell you everything you need to know. Do not believe your senses, spatial disorientation will kill you.

One other thing I used to teach was APT (Attitude Power Trim)

Hi spagiola,
Well... look at this url for this pic, the same posted by jcjeant: http://jacno.com/prov/images/extrait-fdr-cvr.png
jacno.com = Norbert Jacquet.
If one really knows who is this guy, the vitriolically anti-Airbus is obviously an understatement. It looks like he's got few fans around, including here (jcjeant,...), that are pursuing the very same crusade than him by always relaying all his "very informatives and unbiased" positions.
Call them trolls, whatever, but be sure that this won't stop any time soon.

missing vital clues
 

G'day all,

Very fascinating discussion going on here guys [and gals], well done.

At this stage we are obviously missing some vital clues but I would like to pose a couple of questions if I may:

1> Given the data released thus far can we be absolutely certain of complete system and airframe integrity up to impact?

2> Do we know without doubt there was no degradation of other instrumentation including altimeter & AI?

3> Are we in a position yet to completely rule out external factors including extremely volatile air masses?


Cheers

Not according to those in the know.

K.I.S.S.

Why can't it just be simple? All these technical issues are very interesting but are you drilling too deep? I'm not argueing for/against the AB philosophy (for which I am the No 1 supporter; I just love the 320/330), only trying to come to terms with what actually happened.


They lost the pitots and, therefore, airspeed tapes. They had not experienced this before (I have never had a go at this at high level in the sim or the A/c). The PF chose to accept he was overspeeding and pulled the nose up. The auto-trim responded and wound in the nose-up trim. That's what it does. PF is now three minutes from crashing and the captain (probably wisely) didn't exchange seats with either of them (who knows what the turb was at the time?).

Lots of NU trim, at night, no airspeed tapes and three minutes to go.

Would you have done better?

Having spent years watching crews in the sim sorting out problems and noting the time it has taken them, three and a half minutes is not a lot.

Come on guys, for all we know they DID check the GPS/inertial ground speed. The report just does not tell us anything of the sort, just as it does not tell us anything else other than some stuff that does not make any sense at all. So please don't say: why did they not do this or that. We just don't know yet.

Of course a lot of scenarios spring to mind, but none can be verified.
Take for example the 16,5 degree pitch up that someone suggested might be a windshear avoidance reaction. Well, maybe. But you would expect TOGA to be applied in such a situation, which did not happen untill much much later! So much for the windshear scenario.

Did the PF think he was overspeeding, hence the initial pull up? Maybe. But he did not close the thrust levers untill 3 minutes after this pitch up.

So no matter what scenario you (or I) come up with, it does not make ANY sense. We are not given crucial information.



Anyway, to me, just as puzzling is the weather these guys were in and it surprises me there is no more discussion about it. There must have been a lot of water floating around at FL350 to cause the pitots to block. I have never flow commercially in those regions so I can't talk from experience about this. So what I'm asking is, is it known to have icing at those altitudes without entering a CB? Is it common? Is it possible?

Ashling
I believe they had sensory overload and made an incorrect assessment.

I concur whole heartedly with your post.

Unfortunately many will bury their heads in the sand and say it can't happen to me.

What many of you will not have done is to be locked into a decompression chamber and be starved of oxygen as was compulsory in my training. We had to sign a legal waver and had a doctor and medic in the tube with us. The exercises that we performed were to demonstrate our mental degradation at altitude leading to unconsciousness.
Not posing but just some training is more comprehensive than others.


I recently did a very well briefed SIV paragliding emergency course nr Annecy.
During a relatively extreme maneuver (high G) I put it the wrong input although I had 2 way com and was fortunate that I didn't have to throw my reserve.

In my younger days I was known as a bit of a hot shot - I now accept that I am an old man and make lots of mistakes.

Fortunately my last company look after me extremely well and although I still have a valid atpl I haven't needed to work for more than a decade.

I am just happy that I don't have to fly 12 hour night flights through the ITCZ anymore.

bienville...

You found this site and thread and leapt onto it like a young tiger -- a young tiger who hasn't yet learned most of life's lessons, including respect for other tigers. In one of your first posts you insulted a very respected and long-term ppruner who has more wisdom and experience in her little finger than you seem to have in your whole being.

Despite your screams of pilot error, there is nowhere near enough info (publicly) available yet to support any of your assertions. Your simplistic view of what went on up front on that horrid night shows you have little kowledge (or concern) about how the accident chain works in real life.

You yell at ppruners who have been here for years -- and discussing AF447 for years -- presuming to instruct them in how to behave and how to identify trolls.

Saddest thing, though you have been here less than 24 hours, you have succeeded in bringing the level of this thread down dramatically.

JT? You out there?....

Watch out you are being a bit melodramatic! :}
Well said!

Windsheer
 

Out of my league, but could the assumed loss of situational awareness been due to belief of a non-existing wind sheer pulling the plane first up and then down?

I mean, purely from psychological point of view, they just entered a weather front, then loose speed indication and for what ever reason the plane is first rapidly gaining altitude only to loose it soon after. Could this be part of the scenario, making the believed wrong calls a bit easier to comprehend?

cargo ? fuel trim ? both?
 

It went into the deep stall too easily. Should the cause rest only on the THS trim and/or the PF's pitch-up inputs ?

Sorry if I seem rather intuitive than technical...

Thanks, Grizzled
 

Well said.

Many posts discuss the continuous back stick, but nobody replied to my earlier thread re the sustained (30s) full deflection roll input.

I find this very significant - but puzzling - for a number of reasons:

1. I can only suppose that it was in response to an opposite bank, caused by turbulence => the PF was following the AI in roll; why not also in pitch?

2. 30s is an age to hold in a roll demand - an absolute age. If the aircraft wasn't responding (which one supposes it can't have been) then this would to most of us must again re-inforce the conclusion of stalled flight and certainly eliminate a high speed dive.

To me this can only point to extreme disorientation/sensory deception and refusal to believe the instruments. As has been noted before, a/c have been lost due to e.g. a single AH failure; the level of failure here was more serious, was in IMC at night...

More to it than just pilot control inputs
 

I am a pilot and a software engineer. Back in April I posted about a scenario that turns out to be quite similar to what the FDR and CVR describe. In that post I suggested that the pilots would have applied TOGA and full nose up trim. I think it is eminently reasonable, given the pilots' procedures training, practiced in the sim, and their confidence in the FBW system, that the pilots applied nose up control inputs and full power. Don't blame the pilots!

I believe the fundamental cause was that the computers flying the plane used invalid air data inputs (altitude and airspeed) with pilot control inputs that were based on the assumption that the computers had valid air data inputs (altitude and airspeed). I believe the FBW was not designed or tested against this particular set of inputs. That is based on my software engineering expertise and not so much what I know about flying.

Techgeek, what scenario would that be at FL350? No pilot knows of a scnenario where you would pull up violently (7000 feet per minute) at FL350. Mind you, without TOGA! That came much later, with the second stall.

Nope. The AP and AT disconnected exactly as designed and the PF reported "I have the plane".... at this point the aircraft went where the human pointed it. It was pointed up (and left I believe, in response to a right roll). The computers at this point applied the design protections (where possible) based on Alt Law being in effect, but did NOT control the aircraft in any rational sense of the word - that was the bloke moving the joystick around. Unfortunately the human reaction does not seem logical to (most) pilots here and the aircraft ended up in a pretty bad attitude. (Nose high, AoA very out of shape, and minimal airspeed).

Why is this so hard to understand? :ugh:

"...Many posts discuss the continuous back stick..."

Please show where in the BEA report this is stated.......

"Stick" is a misnomer, 'mainly'. The Sidestick has its own character, and inputs communicate with the computer, not the a/c. It is a discussion, not an assumption.

Philosophy and mechanicals are different words, and in their muddying, much distraction.....

It has been a team effort.

One Outsider

I grant you that :)

Richard Feynman - "The Meaning of it All"
 

Feynman, the Nobel Prize winning quantum mechanics guru, wrote about a number of subjects besides physics. He was asked why some students had such difficulty learning to spell certain words. Feynman said the problem wasn't the student, but that the word was not spelled correctly.
Maybe that applies here. If competent, highly trained professionals have difficulty sorting out an already confusing situation in a short period of time, maybe the system is the problem.
Designers need to rethink the interfaces with the pilots, including the complexity of the different levels of degradation. Software engineers need to have a bit less confidence in their ability to produce an infallible product. Time for a major rethink all around, IMHO.

Well my animus for software "engineering" is hereby exhibited - all the fancy laws and protection modes had the precise effect of sending 228 people to the bottom because they dealt with meaningless abstractions, not a real world problem.