Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

AF 447 Thread No. 7

Tech Log The very best in practical technical discussion on the web

AF 447 Thread No. 7

Old 19th Nov 2011, 00:48
  #401 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
Training is a Red Herring. With faulty probes, anomalous solutions, unquantified, and unbriefed, UAS is NOT a Training issue.

It is a BRIEFING ISSUE, and an Mx one.

Smoke, Mirrors.
Lyman is offline  
Old 19th Nov 2011, 00:52
  #402 (permalink)  
 
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 82
Posts: 1,454
Likes: 0
Received 0 Likes on 0 Posts
Thought Experiment

Originally Posted by Clandestino
That technique, while patently wrong, would not be lethal if center of the stirring movements were set around neutral or moved forward as ADIs have shown pitch increasing. Problem is that average input was heavy nose-up. That's not ham-fistedness. That's confusion.
Not disagreeing with you on the confusion.

I doubt that a person could make the control inputs that the PF did with just two fingers. He had to have palmed the stick. The forces appear to be too high otherwise.

Here is the thought experiment:

Suppose a pilot in a stall such as AF447 was in, places the stick full aft for a period of time and the THS runs up to its limit. If he now releases the stick and allows it to center, will the elevator center or remain at its last (30 degree nose up) position?

The answer is likely not as intuitive as it seems.
First, the aircraft is going to be in either g maintenance mode or pitch rate mode of C*. If airspeed is invalid, which mode is effective?

Next, doesn't the computer tradeoff elevator deflection for THS deflection. If the THS is at its limit, there is nothing to tradeoff in exchange for decreasing elevator deflection, is there?

Now overlay the aircraft's oscillatory pitch behavior in the stall over the top of this. What happens now?
Machinbird is offline  
Old 19th Nov 2011, 01:03
  #403 (permalink)  
 
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 82
Posts: 1,454
Likes: 0
Received 0 Likes on 0 Posts
Training is a Red Herring. With faulty probes, anomalous solutions, unquantified, and unbriefed, UAS is NOT a Training issue.
There goes a shoal of Red Herring right now.
Machinbird is offline  
Old 19th Nov 2011, 07:41
  #404 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
iff789 - your last two paragraphs are exceptionally pertinent.

"well, how hard can it be ?" - a few have discovered lately.
BOAC is offline  
Old 19th Nov 2011, 09:34
  #405 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Clandestino
Sidestick command does not order G in absolute terms. It adds G demand to already measured,
When the AP disconnected at 02:10:05, the A/C normal acceleration was 0.85 g (IR#3 p.42), zero V/S and FPA. If the pilot had done nothing, would the EFCS have maintained that level of 'g' (plus or minus variations due to turbulence)?
HazelNuts39 is offline  
Old 19th Nov 2011, 10:31
  #406 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by IF789
I don't think stopping autotrim would have affected this accident. I know you've argued that had the nose gone down they'd have diagnosed the stall, but I'm not so sure. When the a/c did pitch down in stall what was the reaction ? Pull-up, hard. If you're already prepared and briefed for stall in the sim, the nose drop is going to be obvious, but if you think the a/c isn't responding right (which looks like an issue in roll at least even before the stall) and you're pulling back and the a/c suddenly drops the nose, what will you do ?
To trim up in such circumstance must be a pilot's call, not one for the automation.
Let the crew make such a deadly mistake, don't do it for him please.

Trim / AutoTrim: oh yes.
Now we would need to evaluate both item separately ...

But what is the typical problem with trim, what's its MO when it kills ? Looks to me like it's trim-up before stall, autos drop out, trim not managed by crew then contributing to the upset and/or preventing the recovery. Exactly the opposite of 447 - which looks like the odd one out.
Because trimming to the limit of the AP like in AMS or LGW is not a better idea - AP should give back control before reaching such extreme because autotrim itself should not be allowed to go that far.

I note that limits for the 320 electric trim and manual trim are different.
It is apparently not the case for the 330 ?
CONF iture is offline  
Old 19th Nov 2011, 13:22
  #407 (permalink)  
 
Join Date: Jul 2011
Location: Northern Hemisphere
Posts: 195
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by infrequentflyer789
Second:

I don't think stopping autotrim would have affected this accident. I know you've argued that had the nose gone down they'd have diagnosed the stall, but I'm not so sure.
I fail to remember reading a suggestion that "recognizing the stall" was a reason for the autrim to be detrimental. It was rather making the elevators NU even more effective, and the elevators ND less effective, and thus making overall a lot more difficult to get out of the stall.

Someone mentioned the stall incident, and recovery of the Tarom A310 approaching Orly, and pilot action on Manual Trim, and I've read that as a case, in which the pilot recognized early that the angle of the THS will affect how he can get out of the stall, and acted on it.

When the a/c did pitch down in stall what was the reaction ?
Pull-up, hard...
That brings up another problem caused by the behavior of the machine.... They had the Stall Warning sound coming back each time, so under pressure and confusion, they went back to the controls input that silenced that.

Changing something one way to "fix" one accident may make things worse in other cases and end up killing more people.
The cautionary attitude should not overwhelm the technical understanding of the problem.

Originally Posted by CONF iture

I note that limits for the 320 electric trim and manual trim are different.
It is apparently not the case for the 330 ?
I am really curios to see the answer to this. I sense that someone, maybe a group, involved in deciding that, have been asked questions quite a bit by now.

Were the A320 limits on autotrim always that way? Did ever a change been introduced by an upgrade?

Last edited by airtren; 20th Nov 2011 at 12:55.
airtren is offline  
Old 19th Nov 2011, 13:58
  #408 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
Sly, Machinbird. It is not a training issue, per se, insofar as some here are extrapolating holes in ab initio for younger pilots.

STALL recovery needn't be trained in large jets: it is not supposed to occur, and no actual training can be designed for it anyway.

UAS cannot be trained, as in the type experienced lately in Airbus. Witness the meters from AB re: "Don't reconnect AUTOPILOT". "Review STALL procedures," etc. Then, "Wait xxx to reconnect autoflight," etc.

To a large extent, it cannot even be briefed: for 447, it was poorly addressed by both the a/c manufacturer and the line. The Probes issues, though patent, reflect the attitudes present that allowed this crash to happen.

It is NOT established that the PF's initial inputs were NOT an attempt to establish a flight path until speeds returned. How easy to discard this, since everyone wishes to discuss the follow on to this critical event's inception?

Crew were dumped into manual operation abruptly, the jet wanted handling, and it was downhill from there.

Red Herring? How about actively avoiding the discussion to be had re: the onset of Upset?

BEA will be asked to prove the airframe climbed abruptly and STALLED solely as a result of the crew's actions.

I see no such proof. On offer is a gassy spill from Herring merchants.

imo.

Last edited by Lyman; 19th Nov 2011 at 14:10.
Lyman is offline  
Old 19th Nov 2011, 17:25
  #409 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Machinbird in Post #403
Suppose a pilot in a stall such as AF447 was in, places the stick full aft for a period of time and the THS runs up to its limit. If he now releases the stick and allows it to center, will the elevator center or remain at its last (30 degree nose up) position?
Interesting proposition. My initial response is to consider what the aircraft would do longitudinally if in DIRECT LAW.

We know that the CG is forward of the Aerodynamic Center or Neutral Point. That being the case, the Elevators must provide the following:-
(a) The AFT CG position is defined as - 1° of elevator deflection is required to pull a 1g Load Factor, and
(b) The FWD CG position is defined as - the maximum elevator deflection must provide at least the maximum acceptable load factor of 2.5g.

A problem exists in the AF447 stalled situation, i.e. a lack of TAS and faulty CAS, e.g. NCD.

But the aircraft is in ALTERNATE LAW, and the pitch characteristics are the same as NORMAL LAW which seeks a G response to commanded Elevator angle from the SS and as modified by the FCPC. Exactly how the FCPC preconditioning will react with dud air data is the big unknown, but with the aircraft stalled and falling at a nominal 1g, what is known from the DFDR traces is that a reduction of NU Elevator resulted in a reduction in pitch attitude, as would be expected with a CG well FWD of the Neutral Point. Not forgetting the pitching moment resulting from variations in thrust.

So your question -
If he now releases the stick and allows it to center, will the elevator center or remain at its last (30 degree nose up) position?
- seems irrelevant, though I suspect what you were asking is, "If the SS was left to center, what would the THS eventually do?"

Apologies if I have misinterpreted your post.

Last edited by mm43; 19th Nov 2011 at 17:58. Reason: grammar
mm43 is offline  
Old 19th Nov 2011, 18:07
  #410 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by mm43
A problem exists in the AF447 stalled situation, i.e. a lack of TAS and faulty CAS, e.g. NCD.
At the start of the sequence, the speeds were only out from about 02:10:04 to 02:10:36 - just too long to prevent Alternate Law from latching - but interestingly, they come back well before the apex of the zoom climb is reached (The ISIS speeds come back just as the aircraft reaches the apex at about 02:11:15).

From that point the speed is reliable until the stall is so heavily established that stalled air begins to foul the pitot tubes, static ports and AoA vanes - which coincidentally is when the Stall Warning stops after sounding for almost a minute.

Exactly how the FCPC preconditioning will react with dud air data is the big unknown
The major THS movement happened while the FCPCs were getting good air data. The resolution of the graph is pretty lousy, but you can see the beginnings of THS movement following the elevators coming down when it was attempted both times - unfortunately neither attempt was held for long enough to make a difference.

So your question - - seems irrelevant, though I suspect what you were asking is, "If the SS was left to center, what would the THS eventually do?"
My educated guess (and it is a guess, so feel free to take or leave it as such) is that it would have gradually returned to neutral, following the elevator demand. Releasing the stick just after the apex may have been enough for the aircraft to recover itself (being docile in the stall according to the data we have), but within a minute, positive action was required to get them out of it. The nose drops below zero degrees *four times* after the air data goes out for the second (and final) time, but in every case it is hauled back up due to the pitch-up demands coming from the PF's sidestick.
DozyWannabe is offline  
Old 19th Nov 2011, 18:47
  #411 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally posted by DozyWannabe ...
My educated guess (and it is a guess, so feel free to take or leave it as such) is that it would have gradually returned to neutral, following the elevator demand.
My thoughts are similar, though without receiving any positive pitch attitude command, would the THS return to say 4°NU. Remember, the A/P is OFF and the aircraft will follow SS inputs - which it was doing.

The Pitch Attitude / Flight Path Angle and/or V/S would "ring the bells" in this situation, but surely outside the software designer's criteria.
mm43 is offline  
Old 19th Nov 2011, 19:26
  #412 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
A330 - Instructor Support Documentation

I have previously posted a link to the manual, but following recent requests here again is the link to the A330 Instructor Support Manual
mm43 is offline  
Old 19th Nov 2011, 19:46
  #413 (permalink)  
 
Join Date: May 2011
Location: BOQ
Age: 79
Posts: 545
Likes: 0
Received 1 Like on 1 Post
Paradoxical THS

...following the elevator demand.
If the SS is neutral, the command is for 1 G. The aircraft was stabilized at 1 G in the stall. Why for gosh sakes would the THS have to move at all?

If it did, tending to create a load factor of less than 1 G, a situation Mach' alludes to, the elevator would have to then move NU to maintain the neutral, 'hands off' 1 G SS command.

THS may not necessarily follow SS input direction. Example:

In a dive recovery with the SS full back, load factor will be +2.5 G (1.5 over neutral) if the airspeed is 'adequate' to achieve +2.5 G (There will be, additionally, some MLA control surface action). If the aircraft continues to accelerate in the dive recovery, the THS (and possibly elevator) will bias in the ND direction, opposite the SS input, to prevent the G from exceeding 2.5 as the airspeed increases. This not a ND command, just a reduction in NU to prevent over-G but it is movement opposing the command input direction. Depends on airspeed.

Not that you would get there, but I would guess that if full forward stick were providing -1 G, and if the aircraft were accelerating, the THS would bias in a NU direction to prevent exceeding - 1 G (severe discomfort & nausea not withstanding).

(Of course in Direct, instead of Alternate, you can get even more G...in trade for the potential detaching of parts of the aircraft.)
OK465 is offline  
Old 19th Nov 2011, 20:12
  #414 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by mm43
If the SS was left to center, what would the THS eventually do?
After 02:12:30 the SS went through center a number of times, the elevator responded by coming down from -30 to -15 degrees, but the THS did not budge (see IR#3 p.41 for better resolution). FWIW, my educated guess would be that the THS would start to move when the elevator goes past the neutral position (as in Perpignan).

Originally Posted by DozyW
From that point the speed is reliable until the stall is so heavily established that stalled air begins to foul the pitot tubes, static ports and AoA vanes - which coincidentally is when the Stall Warning stops after sounding for almost a minute.
The pitots, static ports and AoA are all on the forward fuselage, not really in 'stalled air'. Some time ago I posted a comparison between the IAS recorded from the ADR's and the CAS calculated from ground speed. The two speeds started to diverge at around 02:11:30. For example at 02:11:43 the values were 100 kIAS vs 133 kCAS.
HazelNuts39 is offline  
Old 19th Nov 2011, 21:42
  #415 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally posted by OK465 ...
Why for gosh sakes would the THS have to move at all?
Right, no G demand equals no movement.

Originally posted by HazelNuts39 ...
... my educated guess would be that the THS would start to move when the elevator goes past the neutral position (as in Perpignan).
Let's mark that answer as correct!

As I see it, while stalled at 1g, the elevator position is a demand for a increase/decrease in G which wont be met until the AoA is such that the wing is flying again. The THS will move to supplement the SS/Elevator demand as long as this G request is not being met. Elevator at 0 equals no G demand.
mm43 is offline  
Old 19th Nov 2011, 22:11
  #416 (permalink)  
 
Join Date: Jun 2009
Location: florida
Age: 81
Posts: 1,611
Received 59 Likes on 19 Posts
Salute!

I go with Doze on this, plus last para of mm's post

As I see it, while stalled at 1g, the elevator position is a demand for a increase/decrease in G which wont be met until the AoA is such that the wing is flying again. The THS will move to supplement the SS/Elevator demand as long as this G request is not being met. Elevator at 0 equals no G demand.
Don't agree with last sentence, as a zero elevator position could result in negative or positive gee depending on airspeed and THS position, otherwise....

From what I read in the FCOM, the THS moves to reduce the requirement to hold back/forward stick. Just like non-FBW planes. So the continued back stick resulted in the THS trimming further and further. I can't find anything concerning THS movement related to measured gee, only stick position.

So I go with Doze that just letting go of the stick may have helped things at the apex of the zoom.
gums is offline  
Old 20th Nov 2011, 04:29
  #417 (permalink)  
 
Join Date: Jul 2009
Location: The land of the Rising Sun
Posts: 187
Likes: 0
Received 0 Likes on 0 Posts
If I misinterpret Lyman I apologise but I rather think he is driving at one of two things. Firstly, a problem with the pitot tubes in which case UAS procedure pertains. Otherwise, failure to follow SOPs - as we know these were not observed. Either way we are left with a human factor issue. The difficult question, why were drills, SOPs and the like ignored? For everything else our shoal of red herrings is now following the trail of clotted cream (and that is a mixed metaphor!).
Old Carthusian is offline  
Old 20th Nov 2011, 14:47
  #418 (permalink)  
 
Join Date: Jan 2008
Location: uk
Posts: 857
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by CONF iture
To trim up in such circumstance must be a pilot's call, not one for the automation.
Let the crew make such a deadly mistake, don't do it for him please.

Now we would need to evaluate both item separately ...
We mostly agree. I think the autotrim discussion probably merits a thread in itself as it goes much wide than this accident.

Because trimming to the limit of the AP like in AMS or LGW is not a better idea - AP should give back control before reaching such extreme because autotrim itself should not be allowed to go that far.
Again agreed. In more general terms I fear the industry may have got itself into a nasty spiral:
  1. more accidents per hour in manual flying therefore we should use more automatics
  2. only times pilots get handed the plane is when George says "something's wrong because we've got to the edge of the envelope, I have limited (no) intelligence and don't know what to do now, so you have control. By the way, that's the stick shaker..."
  3. resulting crash happens with human in control and is then another tick in the "mechanics are better than meat" box
  4. return to (1)
I've thought for a while that it might even be beneficial if George took a mandatory random bathroom break every flight or so, just to ensure the real pilots are kept excrcised. Ecam: "you have control, I'm off down the back for a pee and to chat up the cute FA, back in 15".

Sadly even if it might be good idea over all, whoever implements it will have blood on their hands and a very unsympathetic hearing sooner or later...

I note that limits for the 320 electric trim and manual trim are different.
It is apparently not the case for the 330 ?
I don't know. I'd really like to see more info on this 320 "limit", in particular where implemented and which laws and on what data, but I've found nothing other than the comments on here so far.
infrequentflyer789 is offline  
Old 20th Nov 2011, 19:35
  #419 (permalink)  
 
Join Date: May 2011
Location: BOQ
Age: 79
Posts: 545
Likes: 0
Received 1 Like on 1 Post
I'd really like to see more info on this 320 "limit", in particular where implemented and which laws and on what data, but I've found nothing other than the comments on here so far.
If the 330 SS were released at the apex of the zoom, all else equal, one of two things would have happened:

1. If the existing flight path at the time the SS is released would result in maintaining a speed/AOA in a range out of the stall (speed above, AOA below), then all is well and NO pilot intervention would be required to avoid a stall.

2. If the existing flight path at the time the SS is released would result in a continued decrease in airspeed and a subsequent continued increase in AOA to maintain the flight path, then NU elevator and follow up NU THS motion would occur ‘HANDS OFF’ until which time the pilot would have to intervene to avoid the stall.

This is why I’m a little suspicious of the A320 sim comparison, in which holding significant back stick force was described as a requirement, just to maintain 15 degrees of pitch PRIOR to the stall. Sounds like a low speed stability function of some sort was still in play in the 320?

In ALT2, the A330 would not require any back stick force (i.e. SS could be released to neutral) to hold the commanded pitch attitude/FP, and would do so as long as the aircraft maintained the speed/AOA to do so. Simulations can be ‘right on’ or not, and that’s why the regular evaluations are done using both automated QTG’s, and in the case of stalls (those which WERE done in flight test, i.e. evidently necessary, probably expensive, not terribly dangerous) manual QTG’s are flown to assure the nearest possible fidelity to flight test data that WAS made available.

Further, I would doubt that, once in the stall if the SS was released with the THS at 13 degrees, that the THS would of its own accord just roll back to a position of 3 or 4 degrees, trim positions that correspond to neutralizing the dynamic pressure for a speed range of around 265 KIAS down to around 215 KIAS at 1 G, not 150-180 KIAS and below in a clean config which won’t support a 1 G aero load factor.

If the SS had simply been released in the stall at 40+ AOA and 1 G, what 1 G unstalled flight path does anyone think would have eventually been achieved with the aircraft referenced acceleration already around 1 G and on a current flight path headed steeply down hill with dynamic pressure quite a bit less than that at 215 KIAS and above? ‘Splain it to me please.

Last edited by OK465; 20th Nov 2011 at 19:46. Reason: pitch/FP
OK465 is offline  
Old 20th Nov 2011, 20:01
  #420 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
Without getting too excited, nor eliciting same from partisans, I believe the Pilots were not aware the airframe was STALLED.

Much of what we see in the BEA data supports this. There were insufficient cues to provoke a STALL RECOVERY, unless one certifies that TOGA and PULL qualify. Know the machine? Eh, Carthusian?

Erm, without question, and that means exactly, what? This flight?

This entire thread is peppered with assumptions of superior skills to this crew, which I find frankly repulsive, and quite literally, unsupported.

There is a soft bottom line, of course. To me, continued questions and a resistance to knee jerk pronouncements, plus an unwillingness to eliminate even a remote possibility, is preferred.

airtren, Without THS in the ascent, wouldn't the STALL have precipitated a Break that could have been recognised? Instead of a slide to mush that deprived some critical cueing? Not to mention a sluggish drop of the nose when commanded with FNU THS?
Lyman is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.