Hi,

Familiares de vítimas de voo Rio-Paris querem investimento do governo para novos sensores - Internacional - R7

One would hope that the fact that the Stall Warning was blaring as he made his way to the door and continued to do so for 2 seconds after he entered would have been a clue. In Normal Law one should never hear the stall or overspeed warning.

Hi DozyWannabe,
Do you mean - a clue that the Captain thought they were no longer stalled?
In Normal Law & with the AP engaged, if you are higher than ideal profile, and select Managed Descent from FL390, it's easily possible to trigger the O'Speed Warning.

I'm pretty sure you know what I was saying (the clue is in the question that I answered and quoted)...

Fair enough, noted. However, stall warning should imply that the aircraft is not in Normal Law, can we agree on that at least? That was the question I was answering.

Why not Dozy ?
Again, what are your references ?

Only a few details on the PFD or maybe the ECAM message if it was still displayed.

Hi DozyWannabe,
I agree.
In the cold light of day and in an arm chair - that's an easy question to answer.

However, you believed one should never hear an overspeed warning when in Normal Law - because it says so in the manual - but it is possible.
It also says in FCOM that the stall warning will sound until the angle of attack is reduced to a safe margin - not true.
QRH 2.15 recommends 5 degs pitch with Climb Power. Since Alpha was > 45degs - that wouldn't have worked either.
Nose down activated the stall warning again (bit confusing?).
Altimeters were winding down > 10,000 ft per min - was a computer problem or a breach in the static ducts allowing cabin pressure into the static lines?
W.T.F.I.G.O?

It's taken me longer to write this than they had to recover.

DeFacto
 

It seems clear that whatever the data has or will show, the a/c was handled as if oversped. Now that is an opinion, but the controls and attitudes were indicative of a mitigating plan against Overspeed.

What caused the climb, what casued the determination of the PF that they were fast, etc.?

There will always be questions unless and until BEA release all the pertinent data.

It is all well and good to pitch calumny at one direction or another, but BEA have all the cards, and they aren't talking.

How complete shall the data be to determine cause and effect? Until a likelihood is established? The climb is not unknown, it is is more like a "trait" than an anomaly. So there it is, the CLIMB. I think it very unwise to trust simplistic solutions, make a "software" improvement, and place one's cranium back in the sand.

"Work in Progress" is not something one likes to see in Aviation.......

There is a bottom line, and it has to do with how folks can deduce a chain of events from a recorder with great accuracy, yet the chain of events itself, in real time, is not privy to this exactitude, and suffers the loss. Perhaps Engineering Pathologists should sit in front.

Note the language - "should". References aren't necessary, it's part of the fundamental systems design. In Normal Law protections are supposed to resolve the situation before the warnings trigger. I'm aware that there are exceptions in edge cases (as rudderrudderrat helpfully reminded me).

Fair enough.

If we start debating the merits or otherwise of the stall warning logic *again* then we'll just be going round in circles. This accident identified an edge case which will probably require a re-think of the logic - I'm not questioning that.

All I was doing was answering a question that was asked, namely, would the Captain have had any clue that they were in Alternate Law. I suggested that the stall warning should have been a clue, or at least reason enough to ask what law they were in. I'm not criticising the captain for not following that logic path precisely because the situation was so confusing for him - I'm just suggesting one possible way to work out the current flight control law based on immediate evidence.

* Two texts (the first one is very long, and is shortened here) to enlighten the subject of piloting and training for upsets.


Jacques Rosay chief test pilot « Safety First » magazine

Typically, in cruise at high Mach number and high altitude, at or
close to the maximum recommended FL, there is a small margin
between the actual cruise AoA and the AoA STALL. Hence, in
ALTERNATE or DIRECT LAW, the margin with the AoA SW is
even smaller.
The encounter of turbulence induces quick variations of the AoA.
As a consequence, when the aircraft is flying close to the maximum
recommended altitude, it is not unlikely that turbulence might
induce temporary peaks of AoA going beyond the value of the AoA
SW leading to intermittent onsets of aural SW.
Equally, in similar high FL cruise conditions, in particular at turbulence
speed, if the pilot makes significant longitudinal inputs, it is not unlikely
that it reaches the AoA SW value


The AoA decrease may be obtained indirectly by increasing the speed,
but adding thrust in order to increase the speed leads to an initial adverse
longitudinal effect, which trends to increase further the AoA
It is important to know that if such a thrust increase was applied when
the aircraft is already stalled, the longitudinal effect would bring the
aircraft further into the stall, to a situation possibly unrecoverable.
Conversely, the first effect of reducing the thrust is to reduce the
AoA

UPSET RECOVERY A Test Pilot’s point of view. (FAST N°24) by William Wainwhright Chief Test Pilot Airbus Speaking of stalls and the airline pilot’s training.
For the training managers from American Airlines, Delta, and United,
the only thing necessary was to give an overall industry approval to their
existing programmes; they already worked, because the
many pilots that had undergone training all came out of it with
the same standardised reactions to the standard upsets. For them, this was the
necessary proof that their training programme worked.
Where we differed was in our conviction that there is no such thing as a
standard upset and our reluctance to endorse simplified procedures for recovery
from an upset.
We wanted a general knowledge based approach, as opposed to a rule
based one. For this, after proposing some initial actions, we talk about “additional techniques which may be tried”. This obviously is more difficult to teach.
Even those pilots who do stalls on airtests, as might be done after a heavy
maintenance check, only do them with gentle decelerations, and they recover immediately
without penetrating very far beyond the stalling angle of attack. There
is a world of difference between being
just before, or even just at, the stall, and going aerodynamically well into it.

ON THE USE OF SIMULATORS :
We manufacturers were very concerned over the types of manoeuvres being
flown in simulators and the conclusions that were being drawn from them.
Simulators, like any computer system, are only as good as the data that goes
into them. That means the data package that is given to the simulator manufacturer.
And we test pilots do not deliberately lose control of our aircraft just to
get data for the simulator. And even when that happens, one isolated incident
does not provide much information because of the very complicated
equations that govern dynamic manoeuvres
involving non-linear aerodynamics and inertia effects.
The complete data package includes a part that is drawn from actual flight
tests, a part that uses wind tunnel data, and the rest which is
pure extrapolation.
It should be obvious that firm conclusions
about aircraft behaviour can only be drawn from the parts of the flight envelope
that are based on hard data. This in fact means being not far from the centre
of the flight envelope; the part that is used in normal service. It does not
cover the edges of the envelope. I should also add that most of the data
actually collected in flight is from quasi-static manoeuvres. Thus, dynamic
manoeuvring is not very well represented. In fact, a typical data package
has flight test data for the areas described in Table 1.
In other words, you have reasonable cover up to quite high sideslips and
quite high angles of attack (AOA), but not at the same time. Furthermore, the
matching between aircraft stalling tests and the simulator concentrates mainly
on the longitudinal axis. This means that the simulator model is able to correctly
reproduce the stalling speeds and the pitching behaviour, but fidelity is
not ensured for rolling efficiency (based on a simplified model of wind
tunnel data) or for possible asymmetric stalling of the wings. Also, the range
for one engine inoperative is much less than the range for all engines operating
and linear interpolation is assumed between low and high Mach numbers.
Wind tunnel data goes further. For example, a typical data package would
cover the areas described in table 2. In fact, this is a perfectly adequate
coverage to conduct all normal training needs. But it is insufficient to evaluate
recovery techniques from loss of control incidents. Whereas, the training
managers were all in the habit of demonstrating the handling characteristics
beyond the stall; often telling their trainees that the rudder is far
more effective than aileron and induces less drag and has no
vices! In short, they were developing handling techniques from
simulators that were outside their guaranteed domain.
Simulators can be used for upset training, but the training should be confined
to the normal flight envelope. For example, training should stop at the
stall warning. They are “ virtual” aircraft and they should not be used to develop
techniques at the edges of the flight envelope. This is work for test pilots
and flight test engineers using their knowledge gained from flight testing
the “ real” aircraft.


Vereinigung cockpit, the German syndicate has issued a press release that says that blaming the pilots and only them is far too easy. (On their internet site, press releases)

They are many confusion about different loss of control :
Stall and spin, deepstall, lazy eight, dutch roll, rupture of rudder, are all specific and very different.
Equations are very different. Litterature is not very adapted to describe these different behaviours of the plane.
I know some things about dutch roll, but less about deepstall. Perhaps could a test pilot explain some more about specificity of deepstall which has PERHAPS been the case in AF447.

To the contrary, in Normal Law overspeed protection will resolve the situation only once the warning has been triggered.

No other choice than to be challenged for references as you keep making statements that are in no way part of the fundamental systems design.

Roulishollandais: specificity of deepstall which has PERHAPS been the case in AF447.
Saying that shows that you are honest and open-minded.
In fact I believe it was a perfect example of deep stall. Stable, with only roll excursions. The angle of attack at a very high value, and a stable (awful) rate of descent. To "break" this stable descent, you need flight control efficiency to decrease the angle of attack (with a full "up" THS (or PHR fot the french) was the elevator sufficiently powerful?) and you need to use engine's secondary action (reducing completely will help to decrease the angle of attock)...
But every plane has its idiosyncrasies, and this had never been tried on the A330. I remember, long ago, in Bretigny (flight test center at the time) a French Navy Crusader being tested after two losses following unrecoverable spins. The test pilot used full reheat (post-combustion) to manage decreasing the angle of attack and getting out of the spin. I worked a lot with him later on, and of course, if he had been in the A330...But such pilots are scarce in the airlines nowadays.

I think nobody will disagree that the deactivating stall warning is not a bright idea. It should be noted though that it took a lot of effort to get the a/c that far over the edge.
5 deg /CLB is a worst case interpretation (I'm sure most are aware of this Flight Safety Article? Aviation Troubleshooting: AF447 - Unreliable Speed - by Joelle Barthe, Airbus Engineer ) of the UAS procedure. Had it been applied we wouldn't be having this discussion.

This neat piece of technology has the potential to avoid another AF 447.

Garmin Aera 795 Touchscreen GPS

Of course, it's based on GPS signal and needs to be understood differently to traditional instruments but a few simple rules of thumb allow you to fly any aircraft both accurately* and safely* using this.


*when you have nothing else to help you achieve accuracy and safety.

OK, fair enough and noted (as I said to rudderrudderrat). The protection/warning coupling we're talking about in this case is stall, not overspeed. Several people other than myself have stated that you should not hear the stall warning when in Normal Law.

Given that you've insisted for years that the yoke is an inherently safer system with no evidence to back it up, I guess that makes us even.

Three seconds before loss of Autopilot, the aircraft is climbing at 1000 fpm with her NOSE DOWN four degrees off cruise. five seconds later, the NOSE is passing through 0 degrees and rising.

1000fpm UP and NOSE DOWN is a/p logic for UPDRAFT. If the UPDRAFT disappears, the NOSE will start to rise to lose the indicated and (actual) speed increase. These airspeed anomalies are more than sufficient to waste the ADR's in the eyes of the FC.......

The STALL WARN has a demonstrable failure mode, that we didn't know about.

You are a fair man, and very fluent in systems, DOZE. Do you care to address this possibility of OVERSPEED sans WARN?

The only conclusion I cannot accept as exculpatory is lack of WARN.

Further, these reads and prompts would not be accurate, but sufficient to cause degrade, PF takeover, etc.

It also explains perfectly the ensuing climb, since "crazy speed" cannot be dismissed, especially as an indicated speed.



Sir?

But you still can.
One reason is mentioned by Jacques Rosay here.
Another is the case of a damaged AoA that could trigger a continuous stall warning.

Not too much my style to state that that x is safer than y but please quote me … ?

On the other hand, my position has always been clear :
Sidesticks a la sauce Airbus, is a sure path to waste very valuable information in a multi-crew operation.

The BEA has simply not the latitude to acknowledge such evidence.
Never they would ever mention for the final on 447 that 2 Pilots Monitoring had not the best tool to positively be aware of PF’s inputs.
IMO it is a bare minimum for the Findings.
Beer is on me if they do …

Machinbird, given quality of discussion on this thread, it seems that my attempt at trying to open meaningful discussion on how it's difficult to design, manufacture and install the AoA probe that would work accurately and reliably all throughout the modern jetliner envelope has complettely flopped. Against the arguments "I have moved it by hand easily so it must work below 60kt", "It's not rocket science", "pressure affects both of its sides" and "find an opportunity to handle one" I'm totally powerless. Quite amused, too. Still, I maintain that indignant request for AoA vane to work flawlessly far outside the flight envelope is so unrealistic as to be ludicrous. I can take some comfort form making a big, fat assumption that it stems from ignorance and not agenda.

Anyway, workings of Airbus AoA wane and stall warning system are relevant only in the context of theory that claims the crew which has ignored the stall warning for 54 seconds and after that canceled it's push and pulled when warning sounded again - twice, therefore showing complete inability to comprehend what was the meaning of warning or what is happening to the aeroplane, would miraculously snap out of its confused daze and suddenly react properly only if the warning didn't stop when the aeroplane was where no test pilot dared to take it before. Quite a stretch, isn't it?

Aviation accident investigation entities have obligation to reliably collect facts and analyze them as accurately as possible. Admitting someones baseless pet theories as evidence is not helpful to their aim is therefore not practiced. Moanings on anonymous forums can not be considered to be reliable evidence. Besides, delusion of vocal minority that they're representing the silent majority is as at least as old as written history itself.

I'm still baffled.... why 'blame' the AoA vane?
The often-quoted recorded 60kts IAS was a result of the UAS (blocked pitots), it was NOT the real IAS of the aircraft at the time in question.
A lot of posts here do not seem to have made that easy conclusion....

@ NARVAL

Thank You.

Not only the Crusader has deepstall : the Learjet of MICHEL BAROIN, father of FRANCOIS BAROIN, crashed in a dutch roll, followed by a deep stall... (february 5. 1987). Can the BEA remember that precedent fact ?
Only very very few people, and none in Congo from 1987, could imagine that the pilots would not be able to come out from the dutch roll and try to transform a murder in an accident, and how to organize that (As Dominique Lorentz wrote it in a book (1997 "Une guerre").

Surely the end of the AF447 diving under 45° in the ocean, looks like a deepstall. But "stable" is not the good word when we look BEA's report 3, annexe 4 :
Around 02h10m07s we can see the begining of a lateral autooscillation (page 115), and later around 02h11m40s we can see the begining of a longitudinal autooscillation (page 114). Oscillation does not go with "stability" in the automatician language.

It is strange too that the yaw damper began to oscillate at 02h09m35s, but very slightly. And the loos of IAS happened just after 2s delay in engine deicing : I don't affirm, but it would be well to study the case. Indeed a dutch roll can start on any aircraft when ice comes or goes away from the aircraft.

Another problem is the ability of any stall warning to prevent the deep stall : it is not the wing, but the whole aircraft who has a bad incidence.