PPRuNe Forums

PPRuNe Forums (https://www.pprune.org/)
-   Tech Log (https://www.pprune.org/tech-log-15/)
-   -   AF 447 Thread No. 11 (https://www.pprune.org/tech-log/511119-af-447-thread-no-11-a.html)

Clandestino 18th June 2013 20:17


I don't consider this as "nose is abruptly pitched up"
Neither would I if we were discussing F-16 or similar instead of widebody passenger aeroplane at cruise altitude, though you are right: attitude itself was more problematic than rate of change.


It is, however, entirely consistent with use of the FD as airspeed bleeds off.
"FD order=0" line marks time when vertical mode FD bar is centered. If CM2 chased FD, one would expect FD plot to oscillate around zero, amplitude being inversely proportional to manual flying dexterity.



I'm unable to see this as "Flight director is now firmly below aeroplane symbol"
From about 10:53 to about 11:11. If CM2 tried to follow F/D, it would get him in trouble anyway, he just accelerated the demise.


From this point as the airspeed decays more rapidly, yes, the nose goes up, up, up...as does, if you notice, the FD pitch steering command
It kept demanding 1400 fpm climb as the aeroplane had been already stalled.


PFD picture at 2:11:39
At 2:11:30 pitch is +15, flightpath angle -10, AoA pretty correctly measures about 25° so there is no chance the aeroplane is not already stalled. Nine seconds later it only gets worse.


edit: BTW the PFD picture with the FD well below pitch attitude is from 2:10:27, well before that very critical moment at 2:10:47 when pitching direction is reversed.
Errr... yes. It is there to help visualize what does the amplitude of the green line represent. At 2:10:47 F/D returns, engaged in basic modes HDG & V/S, and synchronizes itself to present RoC, 1400 fpm but rate goes down to 1100 fpm and pitch remains constant therefore nose-up command can be seen up until about 10:55. After the stall warning goes off, stick is pulled, rate goes above 1400, so FD commands nose down (green line gets below 0) to no avail. Again: following FD after the stall warning went off the second time would merely postpone stall for a couple of seconds.

HazelNuts39 18th June 2013 21:21

My version (with apologies to the authors of BEA's Final Report, paragraph 2.1.3.3.1):

The flight director displays, the doubt regarding the relevancy of the aural stall warning and the identification of the possibility of an overspeed situation did not allow the PF to make a correct diagnosis. He therefore implemented a combination of antagonistic actions to respond to both an overspeed situation (reduction in thrust, nose-up inputs) and to a stall situation (application of maximum thrust).

Thus, it seems likely that the flight director exerted an influence.

The flight director displays could have prompted him to command a positive pitch angle, of about 15°. This value is the first option in the UAS procedure for the take-off phase below thrust reduction altitude. It is possible that, even though he did not call it out, the PF had recalled this memorised value and then had clung to this reference without remembering that it was intended for a different flight phase. The conjunction of this remembered value and the flight director displays may have constituted one of the few (and maybe even the only) points of consistency in his general incomprehension of the situation.

Natstrackalpha 18th June 2013 22:43


Seems that all you need for a career is education
and money

Back to the plot. somebody advocated - just leave the aircraft as it is and it would fly stright and level more or less.

If, due to trim condition changes, for example have we got auto trim when auto pilot is out?

Is the aircraft/would the aircraft maintain S&L when in its little mind it thinks it is Overspeeding - even with the A/P out?

Take one aeroplane - set it up for straight and level and trim - aeroplane flies straight and level.

If the nose should rise or ondeed lower then ther would follow and ensuing climb or descent.

If left alone the aircraft in the descent would increase speed and eventually pitch up due to the increase in Lift it would then climb until the speed dropped off again when the nose would lower causing a descent and increase in speed ad-nauseam until (dependant on Trim setting and other things**) everything had equalled out and the aircraft had naturally stabilized in a descent or a climb or straight and level - but** with shifting CofG due to fuel being used at about 2180kgs per hour (in the case of a 320 in the crz, more for a 330) then a climb or descent would ensue. Perhaps then this would explain the climb up to service ceiling? and the descent down to . . see that chart above.

bubbers44 19th June 2013 00:15

I guess if the FD and all that really cool stuff doesn't work you could just lower the nose so you aren't stalled any more like in a J3Cub, P51, F86, F4, Concorde, B747 and all other aircraft certified to fly.

gums 19th June 2013 01:56

Attitude, flight path, AoA, displays
 
Guess we folks from another career field have lost the battle about an AoA indication ( a simple one). Same for a HUD.

Being an aviation dinosaur, I was amazed at the first HUD I saw back in 1971. Easy to use, and really helped to see what the plane was actually doing with respect to the Earth, although we still had the "steam gauges" like attitude indicator, turn needle, altimiter, airspeed indicator, etc. Those were the "primary flight instruments", as the really old dinosaurs insisted on flying with them. In practice, 99% of the pilots used the HUD for almost everything, despite its obvious use with weapon delivery. Sure, we used the steam gauges to correlate with the inertial data on the HUD - the flight path marker ( FPV for the 'bus, I guess). But the HUD had the altitude, speed and vertical velocity ( inertial or baro) and heading displayed. The biggie was the pitch line display. No kidding lines that showed actual angle WRT to the Earth, and not a simple attitude. Put the FPM on the horizon line and you were level. Put it at 2 1/2 degrees down and you had a great approach angle. Put it 1000 feet down the runway and that''s where you were gonna hit unless you flared.

Our AoA was not a separate indicator most of the time. The Navy jets had "indexer" lights beside the gunsight or HUD to provide optimum approach AoA for any gross weight.

Yet I see "professional" pilots here that kiss off a HUD and a simple AoA indication. At the same time, they seem to put up with the main displays that seem very cluttered. Then the chirps and bongs and cavalry charge sounds. Sheesh.

I submit a PR link for a "civilian" HUD. I also think you can look at my leading edge flap failure approach video to see the value of the FPM. And be advised I used power to bring the aircraft vector back up after screwing up, and barely made it to the RWY. AoA bracket was low, and I kept it that way because I still had one pound of roll authority unless I let the AoA increase. Hence I used power to achieve rate of descent and flight path ( rate of climb is proportional to power available minus power required, huh?).

I will guarantee that if the AF447 crew had an inertial-based HUD that the PF would not have continued to pull up to an obscene pitch attitude and allow the speed to decay. The thing would have worked as normal without ANY AIR DATA!!!

Look at a cost-effective HUD that I found for this rant:

HGS-3500 Head-up Guidance System

rant ends...

AlphaZuluRomeo 19th June 2013 10:22


Originally Posted by HazelNuts39 (Post 7898743)
My version (with apologies to the authors of BEA's Final Report, paragraph 2.1.3.3.1):

The flight director displays, the doubt regarding the relevancy of the aural stall warning and the identification of the possibility of an overspeed situation did not allow the PF to make a correct diagnosis. He therefore implemented a combination of antagonistic actions to respond to both an overspeed situation (reduction in thrust, nose-up inputs) and to a stall situation (application of maximum thrust).

Thus, it seems likely that the flight director exerted an influence.

The flight director displays could have prompted him to command a positive pitch angle, of about 15°. This value is the first option in the UAS procedure for the take-off phase below thrust reduction altitude. It is possible that, even though he did not call it out, the PF had recalled this memorised value and then had clung to this reference without remembering that it was intended for a different flight phase. The conjunction of this remembered value and the flight director displays may have constituted one of the few (and maybe even the only) points of consistency in his general incomprehension of the situation.

:D:D:D HazelNuts39

Natstrackalpha 19th June 2013 12:28


Guess we folks from another career field have lost the battle about an AoA indication ( a simple one).
hardly - if you can get virtual strippers to perform on your desktop then you can have a plane on an ECAM or other page - Gulfstreams have Profile Alt -
so tying this picky (stripper idea) up with an accelerometer would give you your 3d ATT. In the design it only has to measure Longitudianal, Lateral, and Neutral or Normal Axis and Bob`s yer auntie. Or `change in the aformentioned long, lat, Neut plane(s) Its not difficult - all you need is a computor programmer a mathmetician and a games developer, throw in a pilot or two and job done. Anyone study Ada?

Anyone from Airbus out there PM me and I`ll have a team down there in a week and we can put it together. Notwithstanding the entire thing to be logical and do-able in which case no one will reply.

CONF iture 19th June 2013 12:44


Originally Posted by HN39
No panic?

  • On perd le contrôle de l'avion là
  • Le problème c'est que j'ai plus de vario là
  • J'ai plus aucune indication
  • J'ai l'impression qu'on a une vitesse de fou non qu'est-ce que vous en pensez ?
Those are comments from someone who is trying to understand, and he is sharing his concerns.

Someone who is panicking would either :
  • not say a single word
  • or would desperately call his mother

HazelNuts39 19th June 2013 13:48

CONF iture,

OK, no panic, just lost.

busTRE 19th June 2013 17:52

NTA


with an accelerometer would give you your 3d ATT.
What information exactly would this show to the crew?

henra 19th June 2013 17:57


Originally Posted by HazelNuts39 (Post 7899718)
CONF iture,

OK, no panic, just lost.

Probably a bit of both.
Completely lost in the first place and then likely creeping up panic preventing them to step back and think about what is really going on, i.e. thinking about the basics. (Pull => cows get smaller, pull more => cows get bigger again)

Natstrackalpha 19th June 2013 18:13


I guess if the FD and all that really cool stuff doesn't work you could just lower the nose so you aren't stalled any more like in a J3Cub, P51, F86, F4, Concorde, B747 and all other aircraft certified to fly.
stoney ground . . . . .

Besides which, sooner or later - someone is going to twig that it had no reliable ATT info and you won`t hear the last of it from me . . .tum ti tum . . . .

What information exactly would this show to the crew?
The attitude of the 3d symbol aeroplane If symbol flying S&L then actual aircraft is flying S&L - with integrated gyros, too, you can also get away from accellerations caused by gravity/centripetal force/inertia and toppling or the need for re-caging - why, you could rig all this up with a . . .laser ring gyro, bring that back into production, its not much.

If you want to play fighter pilots then yea, ok, bring in a HUD - this may even be easier to . . produce than my cartoon type 3D aircraft symbol which acts on all planes/axis. In fact (sorry, am rolling) you could simplify the whole shabuble with a green light on all of the three INS`s:

If L/R/UP/DOWN/Yaw = zero then zero = S&L When S&L (during alignment) a little green light shows on the front of the INS there are five lights 1 up 1 down 1 left and 1 right and 1 in the middle the one in the middle is green. The 1s on the outside (up/dn/left/right) are red - if you get a red light you are not S&L. If you get the green centre one you are.

busTRE 19th June 2013 18:24

NTA


I guess if the FD and all that really cool stuff doesn't work you could just lower the nose so you aren't stalled any more like in a J3Cub, P51, F86, F4, Concorde, B747 and all other aircraft certified to fly.
yeah but what information is your gadget giving them in order to do this?

bubbers44 19th June 2013 18:54

The J3 cub didn't have a battery so no fancy stuff to help but pilots didn't have any problem recovering from a stall. You feel this shaky stall and the stick doesn't do much so you lower it and try to keep is straight. We even had a venturi turn and bank indicator so it was very simple. Now for some people it is very hard with FD's and all the new stuff. Why?

jcjeant 19th June 2013 19:15


Why?
To much info .. to much given choices ?
The human are not always the best .. when they must make choice ... :)

busTRE 19th June 2013 21:02

NTA

The aircraft already has something that displays this info, the FPV. Unfortunately, the information from the giros is subject to drift and is also very eratic in its raw state. For that reason it is damped by baro data, otherwise it would be effectively unuseable. If your ADRs are shot then your baro data is unreliable or erroneous and so the data from the giros becomes unuseable which is why the FPV cannot be used with unreliable baro data (as the crew of 447 had at least at first). Thus, your gadget, whilst a good idea on the face of it would suffer from the same issues and thus be of no use in the situation the were in.

gums 20th June 2013 04:39

Oh man. No names, per J.T.'s observation ( but you can quote me by name anytime and my profile on the entries has my real state of residence, etc.)

The HUD's in the military aircraft since 1968 use the inertial system to display the flight path marker/vector and the pitch lines WRT to the Earth. Early ones had some small drift - think one mile in one hour. Short interval data was prolly a thousandth of a degree off from the real world. The A-7 used a doppler system to help keep the inertial very stable, and you could actually takeoff and align the inertial using the doppler in 15 minutes or so. Then we had GPS. With GPS we could align the inertial in a few minutes.

So negative comments about gyros and such for the HUD/main display flight path should look at the record of reliability and accuracy of such systems since the late 60's. The things work without any air data at all!!!! Think AF447, et al, and loss of air data, but a fully operational inertial system that only uses GPS for some positional data now and then, but displays your actual flight path. Would you keep pulling back stick as the troop did in AF447? I don't think so.

There are some here that have used and maybe are currently using a HUD in the commercial flying bidness. The suckers take about 5 seconds to interpret and feel comfortable. For those who have not used one, try it. The bad news is that the company will try to cram all the magenta line crapola on the display when all you really want is some basic nav guidance and your flight path vector to achieve the desired profile.

Sorry if I continue to rant, but I went thru all the advances in avionics and displays since 1971. I never had to see the FMS stuff and the "magenta line" due to my bidness. I never flew a jet that had auto-throttle, or an otto that flew the climb to altitude with all the heading changes and such. I used otto for heading hold and altitude hold and such, especially on long missions while I snacked on a candy bar or ate a box lunch. Think 13 hours in the chair you are in right now reading this.

We have many tech capabilities to help the crews, but we still have to possess basic flying skills and use all the available tools/displays and such of our planes to get where are trying to go. I have a problem with crew coordination, as most of my time was single-seat. So I defer to those here with all the experience.

So that's my input for today.

Natstrackalpha 20th June 2013 19:38

There is still a world of difference between unreliable speed info due to CADC causing technological problems with the kit due to icing and

screaming up to service ceiling and back a few times . . .

I don`t agree on the need to address, in training, further depths of the system - although - that would not be a bad thing in itself, per se`

We must re-emphasise - "get it off the system!!" Get flying - get it S&L and/or leave the b:mad:r alone, but get it off the system - then, somebody drives it - PF, while PM figres out WIHIH! Not as the usual `in the loop ECAM procedures - but as a distinct and absolute 1 pilot flying and 1 pilot (+any extra pilots present in the crew) figuring out all the inobvious and surreal stuff that has just popped up and ruined your evening.

That way - the 233 ton airliner will sail merrily on its way - until the technological anomolies of the moment have been identified and sorted out. The PF is quite capable of doing the flying AND the nav - right from basic training.

THAT must be a procedure - THAT must, MUST be an SOP and a big fat SOP too!

This gives future crews two distinct advantages - 1., PM assured that the aeroplane is safely flying is not affected by the . . emergency . . ? . .situation and therefoe has an undistracted mind to deal with this new load of . . . questions. Same pilots has complete artistic freedom fot eh same reason - he is NOT therefoe concerned with urgent self preservation. instincts kicking in.

roulishollandais 20th June 2013 20:10

......Very sad ....
EDiT jun 27. :
I was sad after seeing jcjeant banned. I did not knew how to say that. It is difficult to continue discussion when suddenly somebody is missing from his empty armchair and you do not know why, is it due to banning or deleting one's own posts. It starts possible misunderstanding (Natstrackalpha was thinking his post did me sad...). Brainstorming is not perfectly sterilized.
But now welcome back jcjeant !

@gums
As usual your post tells us the good FBW state of art.

Natstrackalpha 20th June 2013 21:46

Telephone call!

Yankee Whisky 20th June 2013 23:35

Controls AF447
 
I am very uneasy with the terms used in the discussions around the Airbus
when it went its way to the bottom of the Atlantic. That is as a power pilot I feel that way. I read, for example, the computer decides the rate of change of the elevator when commanded from the cockpit. And that in 2 seconds nose down stick, the elevator actually travels from 30 degr nose up to 20 degr nose up ! Should there not be a "panic button" that, when pressed, makes direct control inputs (or the computerised equivalent) possible.? The aircraft was flying fully nose up elevator (stick input as well) at low speed and, I would have thought that when a pilot pushes fully nose down, the bloody elevator WILL travel fully nose down in QUICK succession.!!!:ugh: In my days of flying I had to use full controls on many occasions, both in normal and in spin or landing round-outs. Why must modern pilots be deprived of this when the sh"t hits the fan, as was the case with AF447 ? Is it possible that we rely too much on statistical probabilities and accept a certain failure rate as normal ? Sure, high speed aircraft and high weights etc play a part, but should the "feel" of flying manually not be exactly that in cases of emergency ? Control loss in the military and civilian aircraft do give a pilot in most cases the option to hit the silk ! Is it possible that (computer) engineers and software experts
have gone too far in taking control of controlling an aircraft ? And test pilots play by the company book so as to keep their well paying jobs ? Just asking.................;)

HazelNuts39 21st June 2013 07:41


Originally Posted by Yankee Whisky
And test pilots play by the company book so as to keep their well paying jobs ? Just asking.................

The test pilots that I know don't fit your description. They chose their career out of love for flying, and enjoy being members of a team that develops these marvelous machines.

That applies also to the authority test pilots who examine an airplane presented for certification, although they do their work from a different perspective. D.P. Davies explains it nicely in his book HTBJ.

Owain Glyndwr 21st June 2013 09:49


The test pilots that I know don't fit your description. They chose their career out of love for flying, and enjoy being members of a team that develops these marvelous machines.
I'll second that HN39. All those I worked with, besides being fine pilots, were quick to criticise anything they thought was not right and were certainly not "company men"

bubbers44 21st June 2013 11:59

I agree, I will have the pleasure to see Bob Hoover and Chuck Yeager at Reno again in September at the Reno Air Races. They are proof that test pilots are not, yes, company pilots but love what they do.

CONF iture 21st June 2013 15:25


Originally Posted by busTRE
Yeah, I said an airbus CAN stall in normal law!

You mention the case where the EFCS could be momentarily outpaced by an exceptional windshear.
I do mention the case where the system can be duped by exceptional conditions when inputs silently lie to that EFCS.


Originally Posted by Clandestino
Parbleu! Wasn't there any indication in cockpit something wasn't quite right?

Not much for a crew who would have to apply a GPWS procedure in such conditions - Crash as a sure ending.

CONF iture 21st June 2013 15:33


Originally Posted by Clandestino
Now, can we accept that alpha probes need to have some airflow to work reliably or do we keep on harping that they have to work perfectly from zero kt to Mmo, which is quite a wishful thinking?

If 10000 feet/min is not some airflow, what is it ?


Fictional software. As developed and certified - can't.
Then why the OEB disappeared for some aircrafts ... ?


Not necessarily just because it's a FBW of different flavour. Part of it is certainly attributable to her price tag combined with date of service entry - she just didn't trickle down yet to lower tiers operators.
The safety record for both manufactures is not in question, but the accumulation of events regarding the Airbus FBW and its protections is real.

Lonewolf_50 24th June 2013 16:32

gums & USMCprobe: while the FPV points you both raise were of interest, the FPV equivalent (FD) wasn't working in that A330 when airspeed went unreliable. (That is how I understand the system. If I misunderstand, apologies for adding to the noise)..
Per HN39,

I've no experience with FD's but note that they were mostly unavailable in the first 40 seconds of the zoom-climb, and locked on to 6000 fpm V/S then 1400 fpm only after the pilot commanded those RoC's apparently independently of the FD's.
Comment for HN39:

I just can't see any reason for him to zoom-climb to seek the limit of envelope protection.
Even the pilot in the left seat seemed concerned by the right seat pilot' climb. (Per CVR in report)
AZR:

OK, thanks for clarifying that for me. I agree that a law reversion is "something more" to take into account for the crew, but I believe that the advantages of normal+alternate+direct laws are more important than the disadvantages (including the quite possible but not proven 'mode confusion').
So long as the pilots both know how the system works, and have practiced it (see BOAC's "properly trained" point) the law degradation slope makes sense (to me) when one looks at the system as a whole. The training point, the training objecive, would hopefully include knowing your systems and sub systems well, and be able to apply the following trouble shooting approach:
"If this is what's wrong, and these lights/alerts are going off, this is what the plane is doing or about to do, or not doing ... "
Being unable to determine "what's it doing" based on "this is what's wrong" opens a crew up for errors in remedy for a malfunction. (True for a lot of different aircraft and situations)
NTA:

If they had an attitude indicator. why are we here?
If the A/H was operational, then, why did they not look at it?
Ever since the release of the FDR info during the interim report era, that question has been foremost. The answer lies in the realm of post hoc mind reading, or best guess speculation.
We aren't ever going to know, but we can infer a breakdown of the instrument scan of both the pilot at the controls, and his co pilot.
One of the better guesses is that PF was following the FD, at least for a while, which he didn't realize was going wrong ... which it will do when airspeed goes all wrong, as at the initiation of the event.
henra:

A/H that is called PFD.
There is also not the slightest indication it didn't work correctly. Please don't let us go back to that tinfoil hat discussion.
It just seems what they saw didn't ring a bell?
Might this lead one to label the event as being the result of "confusion" rather than "mode confusion" ... I'd better duck ... :}
Barking:

Can anyone explain why Monsieur Le Capitain did not notice an unusual deck angle during his sprint to the flight deck? Any of us who've walked in the aircraft during initial climb, especially on empty positioning sectors as pax, have been aware of the deck angle and the extra effort to walk uphill.
The "uphill walk" might have informed his initial verbal offering of "What are you doing" as he arrived on scene in the cockpit.
jejeant:

I have always suspected that the captain was not in the rest accommodation
So, where was he -- in the lavatory?
HN39:

The flight director displays could have prompted him to command a positive pitch angle, of about 15°.
This value is the first option in the UAS procedure for the take-off phase below thrust reduction altitude.
It is possible that, even though he did not call it out, the PF had recalled this memorised value and then had clung to this reference without remembering that it was intended for a different flight phase.
The conjunction of this remembered value and the flight director displays may have constituted one of the few (and maybe even the only) points of
consistency in his general incomprehension of the situation.
Back to an early discussion on training, recency of training, and BOAC's point on "properly trained" pilots. Seems a decent estimation.

xcitation 25th June 2013 22:23


So, where was he -- in the lavatory?
Or... wasn't the FO's wife on the flight - would it be polite to say hello and chat whilst hubby is PIC?

For me the PNF left seat telling PF to stop climbing several times was the probably the most puzzling data point.
If the guy next to me is repeatedly non-responsive and executing a dangerous change in flight level then it ones duty "I have control" and press priority button, and I expect some confirmation in response "roger, you have control" or whatever the French equivalent is.
It was clearly time to put their game face on and switch from sloppy banter chat to professional communications.
Who knows maybe they were impaired by fumes/gas - this is not a judgement just an observation.

bubbers44 26th June 2013 04:44

Yes, a pilot monitoring is supposed to make sure the aircraft is handled properly. He didn't so everybody died. The PF did everything wrong and pulled up into a full stall with PM letting him. Unless airlines get better FO's this will continue to happen. The captain is required to take his rest break so make sure the two pilots up front know what they are doing. These two didn't have a clue because they were autopilot monitors and couldn't hand fly their aircraft, only the captain could who was on his break.

mm43 26th June 2013 09:30


Yes, a pilot monitoring is supposed to make sure the aircraft is handled properly. He didn't so everybody died.
In the case of this particular PNF and PF, I suggest (as others may have done) that there was a marked personality disparity between them. From what I have been able to deduce over the past 4 years is that the PF was of a dominating personality, and this factor is possibly more involved than any other (except lack of CRM) in the outcome of the flight.

Over in the R&N Memmingen thread a couple of days ago, Yankey Whiskey made mention of "gliding experience" being possibly a prerequisite in preparing pilots in how to handle the 'too fast and high' events. I responded with reference to this AF447 event by pointing out that this PF did have gliding experience, but take heed of what he did. Neither did the PNF do what you have suggested he should have.

Both posts were deleted.

Their presence in that thread was likely at odds with an otherwise seemingly rational discussion on CRM and how it should be handled in an approach to a possible TOGA situation.

Edit:: The last two paragraphs are not factual! My recollection of when I made the post was somewhat confused by something ...:\

NeoFit 26th June 2013 11:09

mm43

Both posts were deleted.
Not at all!
Still here (#68 & 70)

xcitation 26th June 2013 15:23


I responded with reference to this AF447 event by pointing out that this PF did have gliding experience, but take heed of what he did.
Good point, it is a tragic irony that had the PF frozen and not touched the controls there would probably have been a better outcome.

Lonewolf_50 26th June 2013 18:21


and not touched the controls
Hmmm, I don't think that's what the "do nothing" proponents are suggesting.

bubbers44 26th June 2013 19:23

If you touch the controls do something positive to fix the situation. Since they forced a full stall they would have been better off both leaving the cockpit and waking the captain to fix things. At least when he arrived at the cockpit he would have something recoverable to deal with instead of what they left him with in the last minutes.

Machinbird 27th June 2013 00:14


If you touch the controls do something positive to fix the situation. Since they forced a full stall they would have been better off both leaving the cockpit and waking the captain to fix things. At least when he arrived at the cockpit he would have something recoverable to deal with instead of what they left him with in the last minutes.
Since AP had dropped out together with the autothrottles when the air data went south, what would have prevented the aircraft from rolling over onto its back, particularly since it had a significant tendency to roll to the right in its ALT2B mode?

Personally I think that PM was aware of PF's efforts at keeping the wings level, and did not feel confident enough in his own abilities to take over faced with his lack of understanding as to why the aircraft was so goosey in roll.

The initial significant roll oscillation stands out as an unusual situation not described (to the best of my knowledge) by the other aircraft that had experienced multiple ADR failures due to pitot icing. The limited training these two copilots received did not prepare them for flight in ALT2B Law.

roulishollandais 27th June 2013 00:19


Originally Posted by Yankee Whisky
The aircraft was flying fully nose up elevator(stick input as well) at low speed and, I would have thought that when a pilot pushes fullynosedown,thebloody elevator WILL travelfullynose down in QUICK succession.!!! In my days of flying I had to use full controls on many occasions, both in normal and in spin or landing round-outs. PWhy must modern pilots be deprived of this when the sh"t hits the fan, as was the case with AF447 ?

It loooks like Airbus tried to copy in the 80's the F-16 flight management system with a g feedback. Which information had they? Who gave the information? Was is public? (The flight law are still under industrial secret) Were spies at work?
Despite Airbus is using the word "protection" against excessive g (2.5!!!) I suspect the idea was to increase passengers comfort in LIGHT TURBULENCE holding g to 1... In the 80's automation fashion started to have interest to limit sound or vibration with "bruit blanc".
Using the feedback to maintain g to 1 was such a path to realize that goal : limit oscillation of the airliner. (In which point of the plane is a second problem...)

But when the system is piloting z`` with feedback you have to use the stick not to control the elevator but the tendance of the elevator (Airbus S/S action)and you have to integrate the signal in the direct path. The pitch trim does the same work to maintain g=1, not the speed, and is "protecting" (limiting) against quick full nose down.(not a god idea!)

xcitation 27th June 2013 15:16


what would have prevented the aircraft from rolling over onto its back, particularly since it had a significant tendency to roll to the right in its ALT2B mode?
Data appears to favor pilot induced oscillations and as often stated on this thread "making mayonaise".

http://www.bea.aero/fr/enquetes/vol....gure.64.en.png

Roll attitude: the pink line in chart above shows no roll over - only constant roll approx 10-15 deg. Now compare pink to red and blue and you see oscillations appear to be mostly pilot induced. I suggest that the airbus damps the roll disturbances very well on its own probably because of its mass and drag in the roll axis.

Turbine D 27th June 2013 15:50

From the AF 447 Archives
 

Original quote posted by Lonewolf_50
Hmmm, I don't think that's what the "do nothing" proponents are suggesting.
Looking back in the archives, the "do nothing" came from a PJ2 posting:

Original posting by PJ2: I wrote on July 30th, 2009 that the correct response was to "do nothing". I don't claim any special prescience...it's just the logical thing to do when in cruise. When this entire matter was conflated in subsequent discussion, I argued that the memorized items were not only confusing but the entire drill and checklist were poorly written. The "above FL100" memorized item to pitch to 5deg and set CLB thrust was, in my view, intended to cater to high density altitude airfields, not cruise altitudes. But this is not clear in the drill.

There has been plenty of discussion on this item throughout these threads, and I suggest the use of mm43's excellent PPRuNe search tool, to find these discussions.

The BEA press conference comment is, in my view, incorrect. While control would not be lost with an increase of pitch of 2.5deg to an attitude of 5deg, (because the pitch is already 2.5deg in cruise, roughly), the maneouvre does de-stabilize the aircraft when the aircraft is already stabilized in level flight with a good pitch attitude and power setting which were just fine moments before the UAS event. In fact, if one is not trained or accustomed to high altitude handling of a transport aircraft, one may be hunting a great deal with either a CC or an SS to maintain a pitch of 5deg. I just can't see, and never could see this memorized item making any sense whatsoever when in cruise flight.
http://i1166.photobucket.com/albums/...-GnmPntp-L.jpg

PJ2 post continued:
At cruise altitudes, I have argued that this was no emergency at all, it was an abnormal which required standard responses as trained, and which required no action other than to get out the QRH checklist for the pitch and power settings. Most here disagreed with this view, citing the "Above FL100 decision-point, but frankly there is no way that setting 5deg pitch is indicated in cruise.

Whether this PF intended to set 5deg or 15deg or something else cannot be factually determined at this time. But the airplane pitched up, and it was held there until the stall while both pilots accepted the trajectory, the pitch attitude and the loss of energy. We can only surmise why, and that, is, I think hindsight territory
Now, perhaps some correction of both roll and nose down tendency needed to be accomplished to return the airplane to the original flight path, but in a different manner than what would be used at low speed and altitude. BTW, the loss of altitude (300 ft. or so) accompanied by roll and nose down was reported in some other UAS incidents.

There were considerable discussions regarding the state of mind of the PF & PNF:

Original posting by PJ2
Quote:
Ah, but was it surprise or "startle" which affected the PF, or was it abject fear?
Either can be argued but in doing so think about what is being said. What business has "abject fear" got in possessing a crew under the quite normal-for-the ITCZ circumstances faced that night and a system failure which, while disconcerting and even initially confusing, was neither an emergency nor as I say a massive failure of engine or aircraft structure? Why abject fear? Startle yes, but then an automatic response? My question is, what must be assumed for such an answer to be rational and in keeping with all those goals we hold dear?

A rapid decompression would certainly startle someone!, and the drill is there to do. Here, there was nothing to indicate a requirement for immediate action.

I want to be careful here not to portray this up as "judge-and-jury" discussion - it is not. It is an attempt by one experienced captain among many here, to keep the original question open rather than "answering" it with, "it was 'startle' factor, and therefore we need to train that out of pilots".

My point is, surprise notwithstanding, (been there a few times...with adrenaline), the intent of thorough training and flying transport aircraft with experienced crews is discipline when things go wrong, reversion to known responses and effective crew communication and why that didn't occur here. The BEA Report goes as far as it might in my view in answering this question but it isn't a complete answer. I think that is for the industry at large to answer, as per the larger discussion concerning automation and "wither airmanship?" I'm not dismissing the startle response - that would be silly. I am asking for some careful thought before assigning it.
In summary, the best response to an event such as UAS by a flight crew is this:

Original posting by PJ2
The "better way" is in how the response was made right from the start of the event. The SOPs do not change significantly from type to type. The trained crew response to an emergency or an abnormality is, with minor variations throughout the industry, very clear and very specific:

- Take control of the aircraft, (usually the PF) and ensure stable flight and navigation;
- Communicate: PF announces the emergency/abnormality so the other crew member, (PM), is aware and shifts thinking and priorites to the emergency/abnormality, (where applicable, PF takes radio communications responsibilities);
- PF calls for the "ECAM Actions" first, then the QRH checklist where applicable;
- PM executes/completes ECAM actions, clearing messages as they are completed;
- When finished, PM calls "ECAM Actions Complete" and calls up the Status Page for aircraft and system condition review;
- When aircraft/system Status is reviewed, PF calls for any applicable QRH checklists. For example, sometimes landing data requires modification;
- Secure the aircraft for continued flight or diversion;
- Communicate with ATC, F/A's (through their leader) and company dispatch;
- Passenger announcement, if required;
- Monitor changes in aircraft performance, fuel situation etc, as required.
Sadly, none of the "better way" happened as related to AF 447.

Hope the above is helpful and appropriate in the current ongoing discussions...

TD

henra 27th June 2013 18:22


Originally Posted by Machinbird (Post 7911527)
Since AP had dropped out together with the autothrottles when the air data went south, what would have prevented the aircraft from rolling over onto its back, particularly since it had a significant tendency to roll to the right in its ALT2B mode?

Would be interesting to know, what would have happened. Personally I heavily doubt the A330 would have turned onto its back.

It is not a Pitts or an Extra.
I don't think it will bank by itself much over 30°. Maybe 45°. With increasing bank angle you get sideslip. In that conditon dihedral and winglets will start creating an opposing force.
I would expect a spiralling decent. Mmo could potentially become an issue in such a Situation but drag also increases drastically when Approach Mmo.

I'm with bubbers44 on this one. Chances are he would have found a controllable yet maybe banking/descending aircraft.

xcitation 27th June 2013 20:36

Auto Call
 
Would it be easy to have the a/c auto alert an emergency operations center of a stalled a/c at cruise altitude?
Ground support would then call the flight deck using the sat phone to establish a dialogue whilst reviewing the basic flight data via telemetry. There would be a lag of a couple of seconds in the comms however it might restore calm to the flight deck and remind them of SOP.
This would not require any new equipment or expense, probably just a simple software tweak and setting up an emergency ops center. UAVs and remotely piloted a/c require far more sophisticated software.
Then again perhaps the lawyers have blocked this already because of the liability.:ugh:


All times are GMT. The time now is 14:43.


Copyright © 2026 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.