Garage Years

Do you have access to the traces of the autopilot's inputs prior to its loss to the crew? Just a question. (A comparison, an overlay, might help give some scale to PF's inputs?)

Is there a slight possibility that the PF may have been reacting to airframe positions, including ones he had created?

Mr. Optimistic

Could the Pitot #1 (Captain's) have been inop for the duration? If so, why?

TyroPicard

If so, then there is friction, perhaps fatal friction, in the cockpit? PNF's call for Captain's return, was that a call for his Authority as well as his skill? Was PNF too timid to overrule the PF? (perhaps?).......

Hi,

takata
And why he can't be irritated cause the captain not come more quickly on the flight deck ?
He make not one .. but some calls .. so this show his impatience for see the captain on the flight deck
We can argue for years only on those words .. but I think it's not significant words for the main event ...

Hi,

AFAIK .. you can see the AP commands in the graphics in the report N°3 (longitudinal and lateral axis commands and surfaces controls responses are shown)
The graphics timeline begin before the AP tell goodbye

A bit of punctuation in the BEA report would have helped, but I'm aware that's difficult to clearly define from a colloquial discussion on a CVR tape...

"Là je sais pas. Là, ça descend"
Expanded and interpreted: "From what I see, I don't know. As far as I can see, we're going down."

"Là je sais pas là. Ça descend"
Expanded and interpreted: "There I haven't got a clue" (repeating the first "là" to stress it). "But we're going down".

Just putting a comma or caps would have made the intonation and meaning clearer.

And literally of course "ça descend" = "it's going down", which can refer to a lot of things, even instrument indications.

I doubt we, as outsiders, will ever get to hear the original tape.

And I would expect even the BEA is still battling today to match up the real meaning of those CVR 'cockpit voice messages' with what they think the pilots were seeing.

Only a detail, but IMHO maybe in a full transcript, expletives should not be deleted..... They often indicate a state of mind, or fury, or bafflement, or a first reaction to a totally unexpected event. And the exact intonation often speaks volumes....

For the francophones among here, I'm just thinking of the difference between the short-and-sharp "merde" (as when burning your fingers on the stove) and the long-and-slow "ohhh meeerrrddde" (when the soufflé coming out of the oven has totally collapsed).
I expect my anglophone friends can come up with exactly similar examples.

I gather in the current BEA CVR transcript, expletives have already been deleted. Too bad.

Hi LW_50,

I'd been meaning to ask you about that as it happens, because that's exactly what I said!

Did you read 'pitots' as 'pilots'? (In any case it's my fault for using 'pitots' as a contraction of 'pitot tubes')

First, for what is worth: the French pronouns ("prenomes"), like the nouns they replace, have in French two genres: masculine, or feminine. "il' is the masculine form, so it would translate into the English "he" for a person, or "it" for an object or animal.

In this case, I have NO doubts, that the PNF's ""il vient ou pas" - "is he coming or not?" is "Lonewoolf_50's" interpretation, in terms of referring to the Captain.

Indeed, it seem to show the frustration in regards to the PFs actions, but also with his limitations as well in changing the situation, without the Captain's presence

But, it's quite a long stretch to say that he refers to the "a/c controls response", instead of the Captain. That's something a non fluent French speaker would not know.

The "il descend ou non?" would be first on the list, with others as well, while, "il viens ou non?" would not even be on the list, not only in terms of meaning, but also in the context of an emergency call placed shortly before with the Captain, and in the context of using "descend/monte" relative to the "a/c" and PF actions shortly before.

airtren

@RetiredF4:
 

Elevator:
If a very small deflection of the elevator causes "a lot of g", the efficiency of the elevator is very high.
The aircraft is considered to be very sensitive to maneuver.
The maneuver point is the CG, for which the elevator is infinitely effective.
The CG must obviously be as far forward from the maneuver point as possible.
This distance is defined by a maneuverability criteria, that determines that
"at least 1 ° of elevator deflection is required to pull 1 g load factor".
This condition defines the aft CG limit in terms of maneuverability.
But the CG must not be too far forward.
The maximum elevator deflection must enable at least the maximum
acceptable load factor (e.g. 2.5 g) to be pulled.
This condition defines the forward CG limit in terms of maneuverability.

Maximum elevator deflection and extern load factor:

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

Tanscript without ponctuation and timming (silences) is difficult to read and interpret.
What makes the link is that it seems obvious that the captain is answering the PNF previous questions:
2 h 12 min 13
PNF: - Qu’est-ce que tu en penses... qu’est-ce que tu en penses, qu’est-ce qu’il faut faire ?
2 h 12 min 15 -> 2 h 12 min 19
CAP: - Là... je sais pas là... ça descend...

2 h 12 min 13
PNF: - What do you think... what do you think, what should we do?
2 h 12 min 15 -> 2 h 12 min 19
CAP: - Hmm... no idea up to this point... aircraft is going down... [free interpretation of the meaning]

And the reason why the captain seems so puzzled is to be find on what the aircraft was doing after his arrival in the cockpit (0211:43 - 0212:15) added to the lack of informations reported to him by near panicked F/Os.
He did not find a stable situation at all: pitch, roll, alarms, thrust... everything could look incoherent to him during this time window.

There is a little something of Perpignan here.

It has to be add to the complexity of the situation - Beside the known disparities of indicated airspeed, did it trigger a silent rejection of the ADR 1 anemometric values ?

How confused was the system ?

Well, maybe... maybe not. I personnaly think it refers to the aircraft due to context, which makes perfect sense to me (grammatically it's the same as aircraft, "avion" is masculine).

At a guess 2" ice perhaps?:hmm:

Hello RetiredF4,

This can be a major catch !!! Your question is indeed very important!!!

I still don't have an answer to my question, which is somewhat along the lines of yours:

"Is the actuating of the control surfaces proportional to the duration of a control SS command", which could also be a contributor to the amplifying of control surface action, if the perception of the control surface action by the PF is delayed.

You should know it but you seems to be the one confused, or trying to confuse on purpose. One alpha probe doesn't change a glitch until a second one is drifting, which is obviously not the case. In Perpignan, it takes two fully blocked.

Even for stall alarm, the higher value of three is triggering the alarm. An under-reading probe is a sign of something (ice, probe issue) but not something that could "confuse" the system (when was it confused, by doing what?).

Sorry, airtren, but you obviously don't live in France.
You should have SOME doubts....
In French it's "un avion", so an aircraft is "il".
I've had regular problems explaining this to anglophone firends...
"Why do the French refer to a Concorde as "he" or "it", rather than "she" ?"

That said, you may have a valid point... "il vient ou pas" is more likely to refer to the captain than to the response of the aircraft. I would have expected "ça vient ou pas ?" in the latter case (still too colloquial to translate unequivoqually without the full context)

@Conf iture & Mr. Optimistic:
 

In ALT/DIRECT no Hi AOA protection, most proberly PRIMs don't care about that......
For SW the highest value of the 3 AOA's is the trigger.

FWIW currently an AOA sensor replacement program is active.

Why do you think that system would bother to isolate any compromised ADR channel - they are continuously monitored in real time - beside for avoiding that wrong imputs would be used by the system flight controls?
Alternate law 2 is just doing that; C* law is modified; imputs are treated differently with less gain (meaning that default values are used instead of real air data). There is no proportionality between surface deflection and stick imputs until direct law. Those imputs are translated into load-factor demand, and system will deliver them up to the limits (2.5g/-1 g in clean conf.).

Lonewolf, in the English version of the BEA note, the time it takes for the captain to return to the cockpit from his rest is less than a minute.

While we await the English version, the PNF presumably asking where is the captain is said at 2h 10 min 49.

2 h 10 min 49
PNF: (…) il est où euh ?
Uh, where is he?

The timing suggests three things:

1.) The PNF expected that the movements of the plane would have brought the captain back to the cockpit.

2.) The PNF had already called the captain but the BEA did not include that call in its sequence.

3.) The PNF thought he had called the captain, but had not.

We do not have a transcript of the phrasing used by the PNF to summon the captain back to the cockpit, but I don't think even resting pompiers would get to the pompe floor within a minute after the alarm sounds.

Typical mistake.
1. the sentence is unfinished, "euh" at the end means that PFN missed the word and did not bother to complete it later.
2. "Where is it... uh..."
He is looking at something that disapeared or that he did not found. And it could be a part of the documentation if the captain left with it.
The clues:
a) he is calling the captain right after saying that..
b) when the captain came back, one of his first word is "here it is, use that".

Thanks CJ. Do you remember a few weeks back when everyone just wanted the cvr as it would answer everything ! Can understand the editing, but it would help if BEA explained their policy to what they were showing and not showing. A simple <expletive> mark in the transcript would help show tension without loss of anyone's integrity.

Still don't know what the PNF and PF knew about their initial altitude excursion, think it is important as the failure of the Captain to be made aware of the climb is I think crucial to his failure to appreciate what the situation was. If they didn't know, then their failure to brief him wasn't a separate failure on their parts.

As for AoA sensor #1, well it could be a bit of ice counterbalancing I suppose but I rather expected to be told that it was likely due to an airflow issue owing to its location and the unusual wind vector.

takata and A33Aab
I don't really mind about the color of the REMUS that was used. But if an AoA sensor seemed to be blocked for a while, I do think it is worth a clear mention.
Where is that mention ?

If one alpha probe is blocked and the two remaining vary together, would it change a glitch in the ADR rejection process ?

Whilst I don't necessarily disagree with the thrust of the rest of your post, I actually parsed this as:

"il est où, hein?"

They would sound very similar.

(Caveat: I am not a native speaker, but have been listening to French from an early age, as my father taught it, and was once fluent in it myself.)

2 h 11 min 06
PNF: (…) il vient ou pas

There's been some discussion as to whether "il" refers to the captain or the aircraft (Avion is masculine in French, so from a purely grammatical sense "il" could be l'avion.)

If you read the entire CVR transcript we have, there is no other reference to "il" referring to the aircraft. The three pilots usually say "je" ("I"), "tu" (second person singular "you"), or "on" (which here means "we", as in "on monte" -- "we're climbing"). That is, they always refer to either their own individual actions or to themselves collectively (which includes the aircraft, as they are in it). Nowhere else do they seem to refer to the aircraft as a distinct object. In my experience, this is common. You would say "il" refering to an airplane if you were outside it watching it; you would not usually say "il" if you were inside it.

"il vient" also would be a very unusual way to refer to an aircraft maneuver, unless one was at an airport waiting for the aircraft to arrive, for example. If it were "il monte" or "il descend" or "il se redresse" (he's recovering") or "il pique" (he's diving) it would make more sense, but I cannot think of any maneuver where a pilot would say "il vient".

So my conclusion is that "il" refers to the captain, and not the aircraft.

takata and fyrefli and spagiola, thanks.

Perhaps with luck, a native English speaker will help with the English version of the third interim report, perhaps someone like the NTSB or AAIB representative / liaison to the BEA.

I wondered more or less the same. Although I can perhaps understand the logic of the automation handing over with unreliable data and the desire not to complicate its logic still further, is it not possible to provide a simple emergency cruise "autopilot" that could be manually engaged in such circumstances that simply keeps the wings level and pitch at 5 degrees.

Dozy, it read pilots to me. Perhaps me being far sighted and the PC I was using is to blame. If what you said was "apart from the pitots nothing was wrong" then my response to you was to something you didn't say.

Sorry about that. My, what a waste of bandwidth over that little error. :O

Apologies due to you and Old Carthusian for confusing the issue due to a slight reading problem. :O:=:oh::\:eek:

takata:

Sorry, we are not speaking the same language nor in the same terms.

When the Captain leaves the flight deck, there are two pilots, one in each seat.

WHO IS IN CHARGE?

You can't answer that with "they are both co-pilots." There are expected roles for two pilots in a multi pilot cockpit.

So, when the Captain left, WHO WAS IN CHARGE?

The new rule, which was the end of your post, seems to be

Whomever is now sitting the left seat is in charge until I (the Captain) get back and relieve him.

Correct?

EDIT: to all of you kind folks, thanks for explaining the nuances of French in the reference that I presumed referred to the Captain's return, once called by the PNF. No hablo Francais sehr gut. :\

From something gums had said about how the F-16 waffles and falls in a particular kind of stall, it appears that the A330 has a similar "soft" feel to the stall ... but then, it appears that the nose was being held up, which kept it in a stall. Perhaps the "feeling" wasn't all that pronounced, other than the roll excursions early on, and so what the PNF was seeing and experiencing wasn't as obvious to the folks in the cabin, which included the Captain.

According to the procedures in operation at the time, the Junior F/O, as PF, in the RH seat.

That's also my understanding.

Christiaanj,

I think your post's contribution to the cause of the French language awareness of the non-francophone audience is welcome.

The common good is more important than the individual's so a thorough and thoughtful reading of my post which would have helped, if you're interested in avoiding making invalid assumptions, and saving time and bandwidth, of a preamble and unnecessary explanation of "l'avion" genre to me, is of a lesser importance.

Hm.... maybe a while back, your anglophone friends were also picking on, and/or having fun, to the use of a "feminine" name, to call a masculine object, resulting in the "le Concorde".... ... makes me think of the beautiful "Caravelle"...

Of course, I appreciate your honest reinforcing of the "il vient pas" translation, and agree with "ca viens ou pas? ...." regarding which, I would have been in agreement with Takata's take.... which is still bemusing me.

Spagiola,

You pointed out some aspects that I have not thought about before, but now, when I think about them, switched to French language, make a lot of sense... thanks.

Lonewolf 50

I would think PNF would have been PIC in the Captain's absence. He had by far the greater total hours and hours in type, of the two, IIRC. Although, he had just come off "vacation"-- here another need to examine the French wording... off several weeks vacation, or was it that his mandated turn-around rest had been a little short of the required hours? Did I read correctly that PNF had just completed a [short?] rest in the rest cabin; did the Captain ask if he had gotten enough rest?

Again, PNF was sitting in the LH seat; ie, that otherwise occupied by the Captain. Now given the logistics of moving around in the cockpit this may have been just convenience of the moment. On the other hand, the A/C seems designed to be flown in emergency more conveniently from the LH seat.

Well, regardless of what one thinks of that, there can be not doubt that if you want to have some idea of what the pilot actually flying is seeing on his screens, datawise, then the pilot flying has to be sitting in the LH seat. Of course, we would still not know if the pilot actually flying was actually looking at the particular screen for which we would know the data.

But it does suggest to me there should be a policy that the PIC sit in the LH seat, and that the normal procedure in emergency should be for the PIC to request that control be turned over to him. (I do recognize that there are conditions when not making a change could be better, but this does not seem to have been such here.)

I agree we need to know French and AF custom in this regard, and indeed AB' thoughts on which side is better for control in emergency-- given AB did not treat each side the same. It would certainly be helpful for tenancy of the LH seat to indicate the chain of command, in cases such as this.

One other thought... This minature SS seems in this situation only to have the virtue of saving weight. Wasn't this control orginally devised for situations having many g's, where a pilot could at most only move his fingers? Certainly at 2 g's a skier can support his whole body, and easily move his arms about. --OE

Old Engineer
 

On the other hand in the case of augmented ops with the captain on break that would mean in any emergency automatically turning over control to the co-pilot sitting in the his/her "non natural seat", the Left one....which might not be the best idea ( the aircraft should be designed to be flown with equal "ease" :ooh: from either seat, but a co-pilot's natural habitat is the Right Hand one)

Takata,

Thanks for this answer. It seems the answer to the question posted by RetiredF4.

But will use the opportunity to ask a question of mine. But before I do that, if my sense of the tone of your post is correct, please take it as a very simple, non loaded question.

The range of air speeds was from 275knots down to very low values, which is quite a wide range.

Is one default value correct for such a wide range?

If this part of your post is the answer to my question, then let me make sure!

My question was about "stick input duration", and there is no "duration" in your answer...

So, I will rephrase the question, just in case it was not clear, picking some numbers for the sake of clarity:

Is there a difference between the control surface deflection in response to a 1 second long stick input to a position 2/3 of stop, and the control surface deflection in response to one stick input that is 5 seconds long to the same position 2/3 of stop?

Wiggy
 

Yes, that is a good point. Although it does appear that PNF was issuing instructions to PF which possibly were not being followed. But then, without the tone of the statements, what was being said exactly is a bit unclear, I'll grant. Then too, whether at the time of these comments the A/C had elevator authority to implement the ordered (if so) action is likewise unclear. --OE

You're right, another language issue.... with the classic problem that similar English and French words do NOT necessarily mean the same...
The French word for holidays/vacation is "vacances".
"Vacation" in French refers to a "time on duty", so it's the opposite notion from the English.
Hope that helps.

Interim Report No.1 refers to the crew as a "whole", and indicates that they flew out of Paris on the morning of 28 May 2009 (AF446?).

The passenger list provides the information that the Jnr F/O had his wife onboard. There is also a question mark regarding her possible presence on the FD.

There has been discussion a page or two back which suggested that the Capt actually appointed the PF (RHS) as his deputy before leaving the FD. Also, since the accident (takata reports) that AF policy has been changed and that the Snr F/O will sit in the LHS and assume command when the Capt is not on the FD. This action would tend to confirm that the Jnr F/O was both PF and PIC.

mm43
 

Thanks for bringing me up to speed on the Captain's change-of-command instructions. I read his statement in the French, but this was one of those times when I relied on my own French, even though to me it was an ambiguous statement. My exact thought was, "Did he really say that, and what did he mean?" Well, I'm really short on French idiomatic usage and conversational French, and should have not skipped over some of the explanation from those trying to help. --OE

back to the "technical" stuff
 

- First of all, and to A33Z, what's the document you are using for elevator deflection per "g"? And for the c.g. versus elevator moment arm? etc.

About 95% of the military FBW jets have the all-moving HS, and no "elevator". This is extremely important once past the mach, if not essential ( ask Yeager about his X-1 missions). So your document must be pointed at subsonic designs and heavies. And by the way, several subsonic designs have the all-moving HS/elevator.

As far as aircraft movement about the c.g......... where's the center of aero forces, the mean aero chord (MAC)? There are very good reasons to have the elevator at certain distances from both the c,g, and the MAC. Too close and the plane is 'twitchy", and too far makes the beast sluggish, and so forth.

I also question a minimum of 1 deg of elevator per gee. That seems very sensitive, and unless the control surface movement is biased by dynamic pressure, you can get into trouble real fast. I would prefer a surface deflection related to maximum movement AND maximum gee AND dynamic pressure, as we had in the Viper ( probably Shuttle, but have to ask some friends).

- Secondly. I'll beat the dead horse once again. Why doesn't the system have a "standby gains" to be used by HAL when the pitot system fails or is deemed unreliable. Our primitive FBW system had such, and it worked in just the situation AF447 encountered, and a prominent caution light came on.

"Standby gains" prevent the potential problem another poster has questioned - "what does HAL use for control surface "laws" when airspeed is unreliable or missing? maybe A33Z can answer that?

- Finally, I disagree that the plane was unrecoverable without using a horrendous 30 degree nose down attitude, or more. The 'bus appears to have excellent, if not oustanding directional capabilities, whether in a rudder "law" or basically good aero design. Our little jet took rudder away from us once AoA was above 29 -30 degrees to help avoid a spin. Hence, our deep stall was relatively free of yaw and had little uncommanded roll excursions.

Our problem was the pitch moment with the full nose down stabilizer, yet nose up capability was still there. Sound familiar?

The swept wing beasts don't have the sharp stall break that straight wing planes have. The usual problem at extreme AoA is that drag exceeds available thrust if you maintain the "back" stick pressure Concorde, all the delta designs, Vulcan, Shuttle, and the beat goes on. On all of those, you could get flying again without lowering the nose to a horrendous nose down attitude, while using all available power. I speak from personal experience from my hours in the Deuce long ago.

It seems that AF447 came close to "breaking the stall", i.e. getting thrust greater than drag. Using the full movement of the THS could have been the key.

HN39;
Yes, intriguing indeed. I doubt if it's "polling" in nature, as these values would be in continuous calculation, but "polling", (so to speak...it doesn't quite work that way, but close), does affect flight data due to recording rates-per-second. Between the two, I'm not sure what the ADR "saw". Is the two second apparent delay merely just that? ...the time it took for BITE processes to assess/decide and the first test was Mach No.?

We have only limited resources from which to build theories and create tentative conclusions; it has been my experience that one cannot rely upon the flight data, (meaning, one must take into account parameter recording rates), for these things because of polling rates, whereas, as you observe, Mach, CAS, etc are "quasi-simultaneously" calculated and (one assumes), sent to the "users", (the ADRs) for use.

Here, I'm not sure about the notion of "quasi-simultaneous" - are these calculations truly parallel and "held in memory" until all processes are finished and data can be sent onwards, or are they series and the data is sent as it becomes available, and if so, at what point in the calculation is a decision made, for example, if quasi-simultaneous calculations result in one Mach value of M0.26, does the process stop there and bifurcate into a "can the AFS remain engaged?" loop, or is it more recursive than this, "just to make sure" of data correctness and a decision to disengage is confirmed? From my reading of the AMM, it seems a lot simpler than this, and I wonder if the two-second "delay" is actually a delay or is it the nature of flight data itself?

Hi,

mm43
Thank you for this information :ok:
I miss it in my read of the report N°1
So this make more weird to me the question of the captain to the Jnr F/O if he had the good qualifications for take the duty .... as reported in the report N°3
So for the flight Paris - Rio ... the captain don't bothered of the qualifications .... :sad: or this Jnr F/O was never in charge .. for the Paris - Rio ?

The wife of the PF had gone with him to Brazil because the crew had a long layover in Brazil. As it was also Pentecost weekend, this meant a long holiday weekend. She taught physics in high school. Their two young sons remained in France.

The PF had the most flying hours of the three in the previous 30 days.

Hope that clarifies the circumstances of her trip, which I had mis-interpreted.

As a Frernchman, jc, consider: Mrs. PF may be seated on the FD, in attendance, with her husband as PF. Perhaps a Plum for PF? Captain's question about the licensure was to reassure Captian as to legality? The seating arrangements and Ile de Sein recovery were reported amid mystery as to who sat where, and why the #4 was retrieved.

But......what would be the relevance anyway, if all was 'kosheur'?

Simply this. Ready to be corrected, but I am in a large family with strong representation of Native French. It is considered extremely RUDE to correct someone in his work. It is inexcusable to correct him in front of a woman, especially his wife.

to jcjeant [QUOTE] So for the flight Paris - Rio ... the captain don't bothered of the qualifications .... http://images.ibsrv.net/ibsrv/res/sr...y_dog_eyes.gif or this Jnr F/O was never in charge .. for the Paris - Rio ?

From interim report july 29

 Flying hours:
total: 2,936 on type: 807(9) in the previous six months: 368 hours, 16 landings, 18 take-offs in the previous three months: 191 hours, 7 landings, 8 take-offs in the previous thirty days: 61 hours, 1 landing, 2 take-offs
This pilot had performed five rotations in the South America sector since arriving in the A330/A340 division in 2008, including one to Rio de Janeiro. His oceanic route qualification was valid until 31 May 2010.