BEA 3rd Interim report:
What would the FD command have been between 2:10:47 and 2:11:38?

Yes, I'm not looking for anyone's own research or opinions, as I'm working on a Wikipedia-article. The absolute truth is not the aim, but anything stated need to be verifiable. So, prefer to write less and accurate, basically anything originating from official or otherwise trustworthy sources.

Caygill
 

This is official:

http://www.bea.aero/docspa/2009/f-cp...90601e3.en.pdf

According to the SELECTED VERTICAL SPEED trace on page 111, FD command was for a 1000 feet/min climb.
For that time period the actual V/S has fluctuated between +3000 and -8000 ft/min ... the horizontal FD bar has shown some push commands as well as pull ones.

When FD 1 and 2 disappeared, the airplane was passing through FL350 in descent ... so maybe the PF, and also PNF for an instant, were trying to not let go the initial assigned flight level with a full nose up command ... ?

Wouldn't that depend on the selected AP/FD vertical mode?

Maybe not completely fitting here but there are two facts I am astonished at by the plane and its performance:
Throughout the ordeal it remained stable to a certain extent, just some rolling oscillations that were possible to correct by either the pilot or A/P.
In my opinion this is the result of new building techniques.
Older built planes never had wings that were absolutely identical, there was just too much spread in machining techniques. So they ended up with different lifting values, and most likely also weight differences.
That would have meant the plane will drop a wing and enter a spin when flown like this. MAYBE the crew would then had been able to get it out of its death track, given enough altitude.

That does not mean I am advocating to build wings a-symmetrical....
But it is like our new navigation/airdata systems, You can have a head on between 30 and 40W with planes flying opposite tracks at the same altitude. Something just shy of impossible in the old days.

The other unbelievable performance was of the engines and cowling design.
Delivering full power in thin air at flow angles of over 60 degrees is just mind boggling. No engine manufacturer or test pilot would have envisaged the engine would not stall. Older designs certainly would have stalled, thus forcing the nose down.

When still in my flying days I was cautious in selecting full power during a crosswind T/O. (OK, a sideways flow is possibly less favourable as one coming directly from below, plus the fuse blanking the wind).
But obviously modern engines are less vulnerable to off optimum airflows as we have thought they were.....

Anyway, when reading the whole story I am reluctant to say that I would have acted different as a captain, I would also not have realised, expected nor seen that the PF was giving full aft stick all the time.

I am happy to have reached my retirement age without any issue hitting the headlines... I can only recommend everyone to try this too.....

@HN39:
 

FDR trace (page 107 BEA Interim #3) shows HDG/VS selected all the time,
or is this a wrong conclusion?


http://i474.photobucket.com/albums/r...GVS_TRKFPA.jpg

AB33Zab,

I must admit I know nothing about the AP/FD functioning and am just trying to understand. The FCU vertical control area has an altitude window and a VS / FPA window. The pushbutton toggles the latter window between V/S and FPA. My understanding is that the FD guides toward the selected altitude or the selected VS/FPA, depending on the active AP/FD vertical mode, which in cruise (I think) would normally be "ALT CRZ", unless "V/S or FPA" was specifically selected. Am I wrong?

Double Back
 

"Older built planes never had wings that were absolutely identical, there was just too much spread in machining techniques. So they ended up with different lifting values"

You also could have mentioned that the Airbus fly-by-wire planes, by eliminating flying control cables and pulleys, also eliminate the need for rigging. Within reasonable limits, the fbw doesn't care about small left-right build differences.

Vertical Tailplane Separation?
 

If this has already been reviewed, forgive me.

Is there the remotest possibility that the vertical tailplane separated due to an inappropriate full rudder input, while at FL350? (I believe the fin was recovered some distance from the site where the fuselage and wings lay).

This could account for the unusual oscillations and strange handling of the damaged aircraft. I have no idea whether the rudder inputs and outcomes were analysed on the FDR.

This is not a pet hypothesis, obviously, more a thought. Has this possibility been examined?

Bloggs, you are not forgiven
 

Please do the bare minimum of research before posting.

This has been raised in dozens of posts, and the answer is NO, NO and NO.

It was attached when the plane hit the water.

P.S. the "strange handling" of the aircraft was done by the pilots. Before that the plane was fully flyable, if they could work out how.

HN39:

Since the FD pushbuttons were not selected OFF, each time valid ADR data was recovered, the FD's would have reappeared, and the FD vertical mode at recovery would have defaulted to V/S and the FD vertical command cue would have commanded the vertical speed existing at that time. This V/S would also be displayed in the FCU window, just as if selected.

Deselection of the FD avoids this.

Bird
 

ventus45,
Traces on page 107 and 108 do NOT show any bird selection.

Are you sure about that?

I'm not familiar enough with the FCS mode switching logic on the A330, but I would have expected the FD ENGAGED state to disappear (unlatch) on the occurrence of an ADR F/W - same as the AP 'dropping out' - and a 'new' push on the FD pushbuttton(s) to be required to re-engage the FD(s) and return the FD bars into view. *)

Also, since it appears not to be advisable to re-engage the AP too soon, I would have thought the same would have applied to the FD.... after all, AP and FD are mostly the same system, with the pilot in the 'control loop' in the case of the FD.

*) In "my" days, AP, FD, etc., were engaged by means of magnetically held levers, which indeed literally dropped in case of a fault (with a loud 'clack' added to the symphony orchestra from the CWS). So a positive action was required to re-engage them.

The turn started some 20 secs after the airplane stalled.

@HN39:
 

I lack in depth knowledge too, but it's in the transcript pages:

2 h 10 min 08 The FD 1 and 2 become unavailable.

2 h 10 min 17 The FD 1 and 2 become available again; the active modes are HDG/ALT CRZ*.
The selected heading is 37°.
2 h 10 min 21 The FD 1 and 2 become unavailable.

2 h 10 min 26 The FD 1 and 2 become available again (HDG and V/S modes).
The vertical speed reached the maximum value of 6,900 ft/min.
2 h 10 min 36 The FD 1 and 2 are unavailable.

2 h 10 min 42 The FD 1 and 2 become transitionally available (HDG/VS modes).
The selected heading is 36°.
The vertical speed is 1,900 ft/min and the vertical speed selected is 1,300 ft/min.

2 h 10 min 47 The FD 1 and 2 become available again (modes HDG/ VS).
The selected heading is 34°.
The vertical speed is 1,500 ft/min.
2 h 11 min 40 The FD 1 and 2 become unavailable.

2 h 12 min 52 The FD 1 and 2 become available again (HDG V/S modes).
The selected vertical speed is -6,000 ft/min.
The selected heading is 197°.
2 h 12 min 58 The FD 1 and 2 become unavailable again.

2 h 13 min 57 The FD 1 and 2 become temporarily available again.
The selected vertical speed is 1900 ft/min.
The selected heading is 280°.

A33Zab
 

Which is confusing, but only if AF abnormal procedures for Unreliable Airspeed are not followed (FD 1 and 2 ..... OFF)

@Conf:
 

I had the same conclusion:

Quote:

---

FDR trace (page 107 BEA Interim #3) shows HDG/VS selected all the time,
or is this a wrong conclusion?

---

however:

BEA Interim #3 - 1.16.6.3 Page 47.

The presence of the “FLAG FPV ON PFD CAPT (F/O)” message indicates that TRK-FPA
(Flight Mode Annunciator) mode was selected by the crew during minute 2 h 11, but that the
FPV was unavailable (see interim report 2 for details on the conditions of availability). Based
on a study of the other relevant parameters it may be concluded that the FPV was selected
between 2 h 11 min 48 and 2 h 11 min 54.

BEA Interim #2 - 1.16.2.4.1Page 35/36.
FLAG ON CAPT PFD FPV and FLAG ON F/O PFD FPV (2 h 11)
Symptoms:
Disappearance of the FPV (bird) on the PFDs, Captain and First Officer sides,
and display of the corresponding flag.
Meaning:
This message indicates that the flight path vector (FPV) function is selected but
unavailable. In order to lose completely this function, which is elaborated by
the three IRs, in a way that is compatible with the CFR, one of the following
three conditions must be met for each ADR:
barometric vertical speed higher, as an absolute value, than 20,000 ft/min,
true air speed higher than 599 kt,
measured calibrated airspeed lower than 60 kt.
Once the operating conditions are satisfied again, the FPVs reappear on the
PFD (if TRK/FPA mode is still selected).

@kwateow:
 

The ECAM ADR CHECK PROC drill came only after 02:12:xx and at that time they were already in deep trouble.

When AP disconnected @ 02:10:05 the UAS (iced pitots) must have been lasted NOT longer dan 10 sec. otherwise this ADR ECAM Message would have been triggered.
(This ADR CHECK ECAM message has higher priority (116) above other present level 2 messages,
so would have been in view)

"If two ADR outputs are erroneous, but different, and the remaining ADR is correct, or if all three are erroneous, but different
On both PFD, “SPD LIM” flag is shown, no VLS and no VSW is displayed.
This situation is latched, until a PRIM reset is performed on the ground without any hydraulic
pressure.


the disagree has disappeared."

I think it's possible, even likely, that many pilots would deselect the FD's if they observed a blank black airspeed tape on the PFD, not necessarily only in response to a particular ECAM message.

And if they chose not to, they would, as a minimum, disregard any FD commands until things were sorted out and FMA modes were confirmed as those warranting use of the FD.

CJ: Yes. (I've flown those older type aircraft too. They are different. :))

(BTW: V/S is the default restored initialization for the vertical mode, but if you're within capture parameters for the selected altitude, the mode may jump right to ALT* or ALT CRZ*.)

A33Zab,
It then appears paragraph 1.16.6.3 on Page 47 and VS/FPA mode trace on Page 107-108 could be in better agreement ... ?
The AP/FD vertical mode trace is also dearly missing especially in regard to the nature of the selected vertical speed just before the event. What kind of internal conflict was already taking place ... ?
Haven't we heard very recently how spikes could trigger some very wild behavior ?

- now that's a software challenge! How does a computer work that out?

Caygill

Do you mean "official and so trustworthy" ?

"...now that's a software challenge! How does a computer work that out?"

Look at the preceding ten minutes and decide which set of data is the most sensible and likely.

If the anaesthetised patient has been cruising along with a blood-pressure of 120/80 for the last hour and monitor A suddenly shows 60/20, while monitor B continues to show 120/80 then monitor A is likely to be wrong.

:confused:

I apologize if this has already been sent here: Air France Catastrophe: Victims' Families Propose Grounding All A330s - SPIEGEL ONLINE - News - International

A Strange Anomaly
The Hanover legal practice of Ulrich von Jeinsen, which composed the letter, and the Berlin aviation law expert Elmar Geimulla made mention in the letter that there could be "criminal consequences" should indications of a software error not be thoroughly investigated and another Airbus crashes for the same reason.
Von Jeinsen's motion is primarily based on the expert opinion of Gerhard Hüttig, a professor at the Institute of Aeronautics and Astronautics at the Technical University in Berlin. Just over a year ago, Hüttig recreated the Air France crash in a flight simulator. In the course of the exercise, Hüttig noticed a strange anomaly in the plane's reaction once it goes into a stall. The trimmable horizontal stabilizer, a flap instrumental in keeping the plane on an even keel, automatically adjusted to push the nose of the plane skyward.
Hüttig, a former Airbus pilot himself, and other pilots present for the test were unable to push the nose of the airplane down and thereby escape the stall.
When the BEA released its preliminary report last Friday, Hüttig immediately zeroed in on data relating to the trimmable horizontal stabilizer. During the final minutes of flight AF 447 as it plunged toward the Atlantic, the flap moved from a 3 degree deflection to a 13 degree deflection, almost the maximum possible. "The phenomenon is startlingly similar," he told SPIEGEL.

A Quiet Reaction
Hüttig passed along his simulator findings to Airbus, the European Aviation Safety Agency (EASA) and to BEA. On Oct. 27, 2010, Hüttig received a response from EASA which said that Hüttig's theory was inconsistent with the "current state of knowledge." "We suspect that the anomaly you found originated with the simulator you used in the study rather than with the airplane model A330," the response read.
Hüttig and Jeinsen told SPIEGEL that the data recovered from the wreck of flight AF 447 would now seem to have corroborated the simulator findings. Furthermore, Airbus has quietly reacted to the safety loophole. In a communiqué to airlines, Airbus provided a new version of pilot instructions for dealing with a stall. Furthermore, in the January issue of its internal safety magazine, there is a mention of manually trimming the horizontal stabilizers.

Only blip there, Mac, is that if you look at the PGF Airbus accident, 2 of the 3 ADRs were giving apparently 'normal' readings (having frozen in the cruise) while the third was giving correct and significant readings and warnings but all 3 were 'shut-off' by the logic. There's your software dilemma. Presumably some nice little nurse comes along in your scenario and ideally makes a logical assumption, but in mine along she comes to find all three readings blank.

englishman,
The article was initially posted here in June 2011.

It's somewhat interesting to note that Hüttig claims that no-one present at this simulation was able to get the nose down, while the data shows Bonin applied nose-up inputs almost throughout the accident sequence, which is Airbus' defense against the claim. And this latter info is what we here have essentially subscribed to, as if there was no other earthly possibility.

And yet, and YET.....Pprune posters still insist that the recorded inputs are either incompetent or inexplicable. I am still suspicious of this point. :suspect:

To recap other info that's been presented here: it's been also said that there is no way to reproduce this stall in a simulator, since there's no air data from testing for such. Just trying to save us a few more rotations of the hamster wheel. ;)

@BOAC.
 

"A// If two ADR outputs are erroneous, but different, and the remaining ADR is correct, or if B// all three are erroneous, but different"


Ref A//. equals Ref B//.


You can not leave this to software (software programmers uses the same human logica,
but humans can add parameters - on the fly - to make a more proper selection.
The reason this NAV ADR DISAGREE and associated drill, is presented to the
crew is to prevent software of outvoting the only GOOD source (if any).

If "2 of the 3 ADRs were giving apparently 'normal' readings" as you stated, this NAV ADR DISAGREE
message would not have been presented on the ECAM!

@Mac the Knife:

10 minutes is 'ages' in flight, so much can change in so little time,
10s is nearly the max. to look back as reference.

For FCPC ADR voting principle see Page 54/55 in BEA Interim #1.

This actually happened to me (sort off), anaesthetised patient
(me) suddenly showed 0/0 which was correct & needed immediate intervention. Suppose monitor A had got stuck
at that moment. My training in electronics is a bit out of date,
but in my day "stuck at" was a problem which needed a bit of
attention.

the flying rudder
 

@Mr Bloggs
"Vertical Taiplane Separation"
for instance, see my post in Techlog forum, Thread AF447 N°7, post #729 December 23. 2012. I am waiting a full analysis from the BEA, and not only "NO,NO,NO !".

That is not a conspiracy ; that is a question in the query :O

@roulishollandais:
 

The evidence is in the BEA interim report #2:

The VS was salvaged with the HST jackscrew upper attachments frames 86/87,
The jackscrew itself was located on the site.
Parts of the HST (which fractured on impact) where located in some distance from the site.

If the VS separated in flight there was no structure to take the HST loads anymore and the HST including jackscrew would have been separated too.
Besides that there would have been a load of messages containing Rudder, Elevator and Trim failures and the rear control surfaces FDR traces would not have been recorded until the end.

http://i474.photobucket.com/albums/r.../THSLugs-1.jpg

http://i474.photobucket.com/albums/r...9/HSTDrive.jpg

Structure part in green.

Yes, and there would have been hydraulic failures as well. Didn't happen.

Thank you for your comments, A33Zab and MachinBird :hmm:

Have you seen this?
 

Been posted many times and is quite frankly rubbish.

Actually in my opinion not rubbish
 

Quite a bit of sense there. I'm not going into the A vesus B thing, because I can't be bothered - I've flown both (multiple) types.

However I would add that 'real experience' (despite the type they were flying) would have helped them to overcome the situation they found themselves in.

The originator of the article (A330 Pilot speaks out) excuses the Pilots explaining how the Airbus can lead one to disaster. He is right in some ways in that the Airbus phiosophy can lead to pilots to ignore basic facts (aerodynamic if you like).

These Pilots (or the Pilot in the RH seat) ignored basic pitch cues and flew the aircraft into the water.

That is a fact however you want to stitch it up. The Pilot (or Pilots) ignored the high pitch indication and stalled into the water.

I find it hard to forgive Airbus (or Boeing for that matter) for allowing a stall indication to stop because the the A/C IAS has decreased below a certain amount.



Occams razor here I think.

No real arguments about your post until the sentence quoted above. The manufacturers A or B do NOT expect properly trained commercial pilots to HOLD an A/C into a FULLY developed stall. Especially having given them all the warnings prior to the stall, which you were also alluding to in your comments. So we cannot really blame the manufacturers.

It was not just because the IAS went below 60Kts but the AOA was also outside limits. Flight testing into the fully developed stall is not carried out so they do not know how the interrupted airflow would affect the various readings.

Hamster wheel again!

Despite all the training and all alarms announcing the stall .. manufacturers now know (and I think they knew it before the case of AF447) it is possible that pilots hold an aircraft in a fully developed stall :uhoh:
What will be their proposals to remedy this ... if any .. and so break this hamster wheel ..

There has to be a trade off between safety features and system complexity and with this a certain level of pilot competence is expected. Probabilities feature largely in how systems are designed and what protections are required

Having said that, if the type of deep stall that has occurred here can render the normal stall warning useless due to airspeed < 60kts, would it not be possible to derive a warning from a combination of high rate of descent / Vertical speed together with a high nose up attitude above a pre-determined or abnormal value?

Its possible that incidences of deep stall do not justify this change to the normal warning / protections.