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

AF 447 Thread No. 5

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

AF 447 Thread No. 5

Old 11th Jul 2011, 16:08
  #81 (permalink)  
Posts: n/a
And in so doing establish a commitment to a particular, and artificial, Angle of Incidence. This is fine in cruise, but maneuvering an a/c which has cranked in a preference for one (Pitch)or the other (never both), becomes a challenge in adapting to the response of one airplane depending on a very important factor.

Changing the characteristics of an airfoil has dangers inherent in its "benefits". Neutral stick is no such thing when 16 degrees NU, it only "Logs" neutral, and when the deflection cannot be read, there is your guessing game.

I'll stick with swash, and internals. Maybe a particle separator. For the motor, not the pitot.

In the Traces, I'll be looking for gee. Just like Perpignan, but this time at the beginning of the crash.

Last edited by bearfoil; 11th Jul 2011 at 16:19.
Old 11th Jul 2011, 16:19
  #82 (permalink)  
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 6,749
Received 126 Likes on 78 Posts
Bear, take a look at page 5 here. http://www.kamanaero.com/images/PDFs...2020050727.pdf

I flew for a few years with Kaman's servo flap system. It's very responsive to pilot inputs. You don't need hydraulics(though they are handy, and with hydraulics gone, the collective is tiring to move and hold in position).

I personally prefer it, stick and rudder wise, to the swashplate that needs 3000 lbs of hydraulics to change pitch on rotor blades. (Down side, of course, is more parts and thus more maintenance worries ... so it goes, no free lunch ... )

As to the A330 THS and its teaming with the elevator, is there something to be learned from the UH-60?

In that cockpit, you will always know where your Horizontal Stab is (a FBW surface that moves) because there is a guage that tells you its deflection from the horizontal, on the center console. (That gauge is pretty important, since if you are flying above certain speeds, full down Horizontal Stab will kill you due to a nose pitch that the rotor system can't overcome).

From the various material I have access to, on the A330 can all up your THS on a standard ECAMS page. I am not sure if that display is the default or not.

Food for thought, in re knowing what your airfoils are doing while you are flying. Did the crew know where it's THS was? Interesting question, but the more critical question, beginning with the initial LW/NU move on the side stick, is

Did the crew know where their nose was relative to the horizon? (Particularly before apogee was reached).
Lonewolf_50 is online now  
Old 11th Jul 2011, 16:23
  #83 (permalink)  
Join Date: May 2011
Location: here
Posts: 131
Likes: 0
Received 0 Likes on 0 Posts
I don't quite understand the obsession with the autotrim - clearly this would be a concern if they had gone down with sustained full ND input, fighting a hopeless battle against the evil THS. From what we know this was neither the case nor do they seem to have lost all pitch authority...
Zorin_75 is offline  
Old 11th Jul 2011, 16:26
  #84 (permalink)  
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 6,749
Received 126 Likes on 78 Posts
The concern seems to be in sussing out whether the trim functions in ALT-2 (or in AA Law, which some still suspect was in effect at some point) had an influence on nose pitch attitude in the first half of the event. If no, then solution is in one direction, if yes, then it goes in another direction.

After that, the problem seems to be one of pitch authority: was there enough?
Lonewolf_50 is online now  
Old 11th Jul 2011, 16:56
  #85 (permalink)  
Join Date: Jan 2008
Location: Blighty (Nth. Downs)
Age: 77
Posts: 2,108
Received 4 Likes on 4 Posts
Sidestick, Elevator, and THS

Quote from PJ2:
Your previous post on the THS on thread #4, here, explains the THS operation very well and this one connects that explanation with how the THS likely functioned with AF 447. I think its a reasonable explanation of what occurred to the THS and how, after the initial pitch-up.

Thanks for your kind acknowledgement of my recent beginners' guide to "Primary and Secondary Pitch Control" on the previous AF447 thread, which also attempts to describe very briefly how the basic concepts are applied on Airbus FBW, and their relevance to AF447.

The key to the THS movements are, of course, how much elevator the EFCS is using: you explained that here before I ever did.

In a couple of follow-up posts, including the one you have quoted in your post, I tried to develop the argument to describe part of what may have been happening with pitch control as the aircraft reached its apogee, and immediately after:

Finally, I offered a partial explanation of the problems the PF seems to have been having with roll and pitch control using the sidestick, and how the EFCS may have interpreted what it thought he wanted in pitch.

Quite how the aeroplane eventually established itself in as others have commented a remarkably stable descent at AoA > +40 and pitch ~ +15 remains mysterious, partly because this is uncharted territory in terms of the aerodynamic performance of the fuselage and THS, as well as the wing. The resulting pitch-moments are therefore unclear. In that context, I recommend a look at Meikleour's contributions, the first of which is
here, which relate to a simulator exercise many of us have not done.

The other unclear factor is how the C* pitch-function of the EFCS would treat the invalidation of CAS (IAS) data, when determining the crossover from g-control to pitch-control. As I understand it, that crossover is normally a gradual transition as the airspeed falls below a certain figure in routine flight. Understanding how the UAS affected this will be the other key step in solving the relationship between sidestick position, elevator/THS position, and the achieved trajectory.

As these pieces are on the previous thread, the links provided will enable you to reach them more easily, and I look forward to any comments or objections. If none should be forthcoming, I shall infer that even the most erudite of our contributors find nothing to disagree with in my proposals...
Chris Scott is offline  
Old 11th Jul 2011, 17:09
  #86 (permalink)  
Posts: n/a
Did the PF know his Pitch (attitude) at a/p loss? Of course he did.
It was ND, and needed NU. The a/p quits at ND-9 degrees. That much? Either way, the PF input NU, there is no call to question the decision?
Old 11th Jul 2011, 17:14
  #87 (permalink)  
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi Chris,
But this aircraft was degraded to ALT 2 law: a combination of Direct law for roll (stick-to-aileron/spoiler) and Alternate law for pitch, the latter without any high-AoA protection.
In trying to cope with the roll-control problems described (but not yet explained) by the BEA, it is almost inevitable that the PF was making continuous roll inputs, and probably had the stick in the palm of his hand. This would have made accurate pitch control difficult.
I agree.
It would only have needed a small amount of rudder trim to have levelled the wings (a technique forgotten / not learned by FBW crews). The stick could then have been released and used more normally to control the attitude. I suspect PF was struggling somewhat to maintain wings level (Meikleour previously described how "twitchy" it was in roll in ALT LAW during his loss of airspeed event).
rudderrudderrat is offline  
Old 11th Jul 2011, 17:27
  #88 (permalink)  
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
Chris - I understood that the C* in the AB switched at around 210 kts so I would expect the FCS to be pitch rate driven during the 'climb' and not Nz, especially given the supposed low IAS readings?

Bear - how on earth do you arrive at pitch -9? Is that what you are proposing now?
BOAC is offline  
Old 11th Jul 2011, 17:28
  #89 (permalink)  
Join Date: Jan 2008
Location: Blighty (Nth. Downs)
Age: 77
Posts: 2,108
Received 4 Likes on 4 Posts
View from the P3 jump-seat (continued)


Many thanks for going to the trouble of obtaining and posting excellent photos of the view from the P3 seat in daylight. They confirm my A320 recollections, after a gap of nearly ten years. If in its stowed position, it should be safe to roll the P3 seat forward to the position the photos were taken from, although the pitch of +15 might have made it difficult. Was the captain merely standing?

As you say: although the sidestick can be seen, provided the PF's table is stowed, the console lighting would preclude this at night unless the dome floodlight was on. In any case, its position would be unclear if it was in the palm of his hand.

THS position should not be a problem, provided the observer's eyes were adapted to low-light conditions.
Chris Scott is offline  
Old 11th Jul 2011, 17:33
  #90 (permalink)  
Join Date: Jun 2009
Location: somewhere
Posts: 451
Likes: 0
Received 0 Likes on 0 Posts

Originally postted by Confiture :
would you still trim up past the early signal of the approaching stall ?
Why automation has been doing something you would not ever do yourself ?
IF ... the pilots were confused enough to maitain NU inputs, was it necessary for the automation to help them in their confusion ?

1) how on earth can a "system" allow the THS to go to full 13 up in cruise at 35.000' ?
Because 1 hand on 1 stick, connected to 3 xdcers commanded the 3 FCPC - in ALt 2 - to do so.
(In the absence of ss xdcr, FCPC or pitch system faults, it wasn't a 'flaw' order generated by the system itself.)

2) how on earth can a stall warning be disabaled below 60 Kts to just kick in again when control is being regained (and completely confuse the poor crew) ?
Maybe because flying at 60 kts, 15 pitch and 40 alpha was not accounted for?
Ask B. they quit all the signals at 30Kts, so between 30 and 60 flight seems to be possible with that one!?
A33Zab is offline  
Old 11th Jul 2011, 17:45
  #91 (permalink)  
Join Date: Jan 2008
Location: Blighty (Nth. Downs)
Age: 77
Posts: 2,108
Received 4 Likes on 4 Posts
Yes. That the PF was struggling with roll-control may have overloaded him. Later, at very high AoA, resultant sideslip would add to probe problems, as has been said. At that stage, was the use of aileron/spoiler productive, counter-productive, or irrelevant?
I'm interested in your rudder-trim argument, but don't see how it would counter wing-drop if the latter was caused by turbulence.

The 210kt crossover seems to be a popular figure. I think there must be a gradual blending-out/blending-in process, to make it seamless to the pilot (which it is). But, as I said, what logic is applied when/while airspeeds have been declared invalid?
Chris Scott is offline  
Old 11th Jul 2011, 17:54
  #92 (permalink)  
Posts: n/a

Steady on. -9 degrees is the a/p limit, the a/c Pitch has not been supplied by BEA. Except in + range........ (perhaps ND does not concern?)

Old 11th Jul 2011, 17:57
  #93 (permalink)  
Join Date: Dec 2010
Location: Middle America
Age: 83
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
Good morning Bear,

I do believe one has to look at the two minutes before AP/AT disconnect and at the moment there is no information from time of 2 h 08 min 07 and 2 h 10 min 05. The speed was reduced to about Mach 0.80 and the turn to the left was initiated at the start of this time period. Although the BEA never specifically has said the pitots became clogged with ice crystals (nor should they at this stage of the investigation), it is apparent they are strongly looking at this as the starting point of the total event. I also think the clogging was not one where in one-second the pitots were clear and then poof, in another second they were clogged. It probably happened on a gradual but ever increasing basis over these couple of minutes or so.

So as the pitots began to clog, the sensed airspeed by the computers began to go down. Now I can't imagine the computers observing this degrade in speed sat there and did nothing, but perhaps Takata can provide the computer reasoning and logic in this situation while in AP/AT. I would think they would signal the engines to spool-up and increase the speed to maintain level flight the AP was attempting to do at that time. Unfortunately, the sensed speed degrade was wrong and the aircraft was gaining speed it didn't probably require in level flight. I would assume that when the AT dropped and thrust lock occurred, the thrust was locked at a higher N1 setting than perhaps the PF thought it was at?

I am not going to comment on this aspect as to what the computers were doing or not doing. But rather, some questions about an observation I noted in reading the reports regarding aircraft balance. In the very first BEA Interim Report, dated June 1, 2009, at the time of the beginning of the incident, the aircraft weight was estimated and reported at 205 t and an aft balance between 37.3% & 37.8%, controlled within 0.5% of MAC. The second BEA Interim Report mentioned nothing in this respect.
However, in the BEA Update, dated May 27, 2011, The weight of the aircraft was again reported at around 205 t, but the balance was changed to 29%, or in other words moved forward 8% or so. I thought the aft balance (37.3% - 37.8% reduced drag and improved overall efficiency.
So my questions are: How did this happen? Why did this change happen? Does this have any effect on maintaining pitch either by the automatics or in a manual fly mode?

Just curious...
Turbine D is offline  
Old 11th Jul 2011, 17:59
  #94 (permalink)  
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi Chris,

Correct - it wouldn't work for random turbulence, but only if it was mis-trimmed to begin with (rudder + or slight fuel imbalance + or thrust asymmetry).

"From 2 h 10 min 05 , the autopilot then auto-thrust disengaged and the PF said "I have the controls". The airplane began to roll to the right and the PF made a left nose-up input."

Last edited by rudderrudderrat; 11th Jul 2011 at 18:47. Reason: extra text
rudderrudderrat is offline  
Old 11th Jul 2011, 18:22
  #95 (permalink)  
Posts: n/a

A change of 8 per cent in Weight would be 33,000 pounds. So the 8 percent of CG is a whopper. Unfashionable to change such heavy horses in mid stream. It may represent the entire aft fuel load, as if it were foreward, which it was not. Cheating Fuel is an old trick, and it saves a significant amount of fuel burn? The cost is Stability, orienting the Cl to the Cg is hazardous. The a/c becomes more sensitive in Pitch, obviously.

If, for instance the Cg fell behind enough, Up becomes down, and down Up. PF did hold aft stick all the way down, after all. hmmm..........

I am not real clear on the a/p's authority in Pitch, Roll, and keeping the two in line with each other. To me, if Pitch transits got out of phase with Power, their could be trouble? Insofar as rate, I think Power up when Nose down would be problematic, as would the reverse. If the THS is in there, what is the calculated rate for stopping speed and Pitch excursions that are (or have become) unnecessary?

If it is a rate thing, do the a/p's limits constrict? As in, "Keeping Up"? I still think it is a toss-up which came first, the 'slows', or the 'unreliables'.

I lean to the slows....... leading to hand flight.

Now, will there be some clarity from BEA?
Old 11th Jul 2011, 18:27
  #96 (permalink)  
Join Date: Jan 2008
Location: uk
Posts: 861
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by bearfoil
As the speeds declined, the ship would accommodate the "loss" of speed by increasing power, since energy needs to be added and altitude needs to remain the same, but as the a/c climbed anyway, (the energy was actually just right, the computer found it to be low because of ICE), the Nose would be lowered. This might continue until the a/c was zipping along at its assigned cruise level carrying too much power and actual velocity. Depending again on the rate of uptake, this false trimming would last until the a/c autopilot checked out, unable to control the Nose "hunting for a satisfactory Pitch". (AutoPhugoid?)
Maybe not impossible, but I think the theory falls at the same test as the idea that updrafts caused or accelerated the climb - namely that the kinetic-potential energy then doesn't add up. Those (not me) that did the maths and posted however many pages back, showed clearly that the BEA figures match the a/c trading KE for PE. Your theory has the a/c carrying additional KE over and above what BEA says (real AS > IAS) before the climb - which would give a higher apogee, I believe.

Bottom line. Why Autotrim in dire circumstances. Obviously available, even mandatory, who needs it?
Yet in several other crashes and incidents, loss of autotrim has lead to LOC (and deaths) as pilots failed to manage the trim along with everything else they were handed at short notice.

Autotrim on or off - no right answer. Whichever you choose as a designer, sooner or later someone is going to die in part because it wasn't the other way round.

At the risk of parsing too closely, In the audio where the pilots are noticing "No indications", is it surprise one senses? Or Betrayal?
Given we've only got a transcript and that through translation - who knows.

I also keep coming back to that line though, and wondering what other indications they lost. Did they lose attitude, distrust it, or just not see it ?
infrequentflyer789 is offline  
Old 11th Jul 2011, 18:35
  #97 (permalink)  
Posts: n/a

Perhaps not a smooth climb, then, but one with rolling and porpoising? Both eat up energy quickly. The Rolling we know, and the Pitching we can infer from BEA?
Old 11th Jul 2011, 18:53
  #98 (permalink)  
Join Date: Jun 2009
Location: Paris
Posts: 691
Likes: 0
Received 0 Likes on 0 Posts
Hi Bearfoil,
I really think that you are chasing your own tail all around this thread.
Originally Posted by Bearfoil
Steady on. -9 degrees is the a/p limit, the a/c Pitch has not been supplied by BEA. Except in + range........ (perhaps ND does not concern?)
The BEA supplied some informations in its narative. Those they considered relevant in order to understand roughly the sequence of events, mostly because there is still many data they would have to check further in detail or to derive from those rough recorded data. But, be sure that it should not cover the flight parameters while she was still flying in auto-mode as it would be very easy for them to verify immediately if everything was ok until the point autopilot disconnected (cross-checking all speed, altitude, position, parameters from the various sources will tell you straight away if it is right or not).

At AP disconnection, the only thing they considered worth mentioning was that the aircraft started rolling (meaning she was wings level before), we don't know the rate or amplitude; but, do you really think that they would hide to us something very unusual, just before this point?
More likely, don't you think that she was flying as expected at FL350, 275 kt, wings level, +2.5 deg pitch, THS should be about ~2-3 deg NU?

Some of this is already part of the narative:
At ORARO : flight level 350, Mach 0.82, pitch attitude about 2.5 degrees; airplane around 205 tonnes and 29% MAC; Autopilot 2 and auto-thrust engaged...
Further change is noted:
2 h 08 min 07: slight turn about 12 degrees to the left; turbulence level increased slightly; speed reduced to about Mach 0.80.
2 h 10 min 05 : autopilot then auto-thrust disengaged; airplane began to roll to the right...

Beside, AP disconnection, one not due to protections kicking (High AoA or High Speed) or Failures in relevant systems, then due to AP limits are:
- Aircraft attitude :
* Pitch > +25 or < -13
* Bank > 45
takata is offline  
Old 11th Jul 2011, 19:08
  #99 (permalink)  
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,688
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Turbine D
I also think the clogging was not one where in one-second the pitots were clear and then poof, in another second they were clogged. It probably happened on a gradual but ever increasing basis over these couple of minutes or so.
I wonder on what consideration you base that assumption. On the basis of how a pitot tube 'works', it is not plausible at all. The BEA Update speaks of 'a sharp fall from about 275 kt to 60 knots in the speed displayed on the left PFD, then a few moments later on the ISIS'. The Air Caraibes Memo speaks of 'une diminution tres rapide de la CAS'. On the basis of the ACARS Fault message PROBE PITOT, BEA's Interim no.1 attributes the initiating event to 'a decrease of more than 30 kt in one second of the polled speed value'.
HazelNuts39 is offline  
Old 11th Jul 2011, 19:24
  #100 (permalink)  
Join Date: Jul 2002
Location: UK
Posts: 3,182
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Machinbird
JD-EE, we were temporarily lead astray by forum members who apparently believed that autotrim would not resume control once you made a manual trim input.
That would be me, having misread the documentation I had.

I strongly suspect there is some misinformation adrift in the Airbus community that needs correction. I wonder how BEA will address that?
Again, it was just me - the "Airbus community" were the ones who kindly corrected me via PM.

Originally Posted by Machinbird
The problem with the trim is that it moved to a high aircraft nose up setting without crew awareness.
But all indications suggest that the PF was commanding the elevator and by extension the THS to do just that - it didn't do it on it's own.

It appears that a FBW aircraft requires the pilot to know exactly what mode the aircraft is operating in or else the question arises, "What's it doing now?"
Not true. "What's it doing now" usually applies to modern autopilots (i.e. FMS/FMC), which are not restricted to FBW aircraft (the 757, 767, Classic and NG 737 and 744 have them for starters), nor indeed Airbus. As far as trim goes, all you need to have in the back of your mind regarding modes/laws is that in anything other than Direct Law and below, autotrim is active, so one needs to be wary of large, sustained inputs.

Originally Posted by CONF iture
You are WAY out DW, and should know better ...
Really? How else am I to interpret at least 5 years of posts demanding that Airbus return to interconnected yokes and introduce a big red "Direct Law" button (as in the 777), not to mention continued belief in a conspiracy surrounding AF296 @ Habsheim?

Originally Posted by rudderrudderrat
If you keep taking tactile feed back away from a pilot, then you reduce him/her to a visual clues only input processor. If the FBW computers couldn't cope without valid airspeed information, is it a bit too much to expect a human to do a better job without the benefit of a "conventional aircraft's elevator feel"?
But "tactile feedback" has been artificially driven in every major airliner designed since the late '60s, and indeed is computer-driven in the case of the 777 - so if you're in a situation where you don't trust the aircraft, who's to say that the feedback you're getting is accurate?

Also, the FBW computers coped just fine with the loss of airspeed information. The A/P and A/THR kicked out, and some protections were lost but ultimately the pilots had a controllable aircraft.

Originally Posted by JD-EE
And I believe we are in agreement with gums about the word "protections." The word sets up the wrong mindset leading to people getting reckless and cavalier about deadly serious subjects.
The problem is that "limits" (gums' suggested term) does not adequately describe what the system will do for you - for example, the Alpha Max/AoA protection will spool the engines to full thrust in the case of a sudden nose-up input - that's not a limit, it is - for want of a better term - a "protection". Also, your comment suggests that you believe some consider the existence of protections and their description as such might lead to a false sense of security - but from what I've learned from this thread, conscientious FBW Airbus pilots seem well aware of the limitations of said protections and the modes/situations in which they will not function.
DozyWannabe is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell or Share My Personal Information

Copyright © 2023 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.