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 25th Nov 2011, 11:03
  #501 (permalink)  
 
Join Date: Mar 2002
Location: Florida
Posts: 4,569
Likes: 0
Received 1 Like on 1 Post
mm43

re: post #510

the melting of this ice occurs in a relatively short time.
please define the boundaries for this relatively short time.

I doubt that the reader knows what it is relative to Micoseconds, seconds or minutes etc.
lomapaseo is offline  
Old 25th Nov 2011, 11:08
  #502 (permalink)  
 
Join Date: Oct 2005
Location: 45 yards from a tropical beach
Posts: 1,103
Likes: 0
Received 0 Likes on 0 Posts
Dear Dozy,

I agree with you about the speedbrakes, in that they should play no part in any stall recovery, other than to ensure that they are retracted.

However, I disagree about the stall being irrecoverable. In modern airliners, an irrecoverable stall (the oft misquoted 'Deep Stall') is purely a feature of 'T tailed' aircraft, hence the certification requirement for a Stick Pusher on such types.

A conventional, low tailplane type such as the A330 should never enter an irrecoverable stall. In the case of AF 447, the self-inflicted stall should have been recoverable by pushing the sidestick forward, then helping the desired pitch change with judicious use of nose-down trim. Depending on the altitude, recovery should be accomplished with no more than 3,000 feet of height loss. (IAS/TAS relationship + momentum.)

For the thinking Airbus pilot, on seeing the amber crosses on the Flight Director, which indicate reversion to Alternate Law, the first thought should be:
"Loss of protections - especially Stall Protection."

The next action (CRM) should be to pass that warning to the PNF, with the rider "Watch my back!" Simply, basic Airmanship.
Neptunus Rex is offline  
Old 25th Nov 2011, 12:01
  #503 (permalink)  
 
Join Date: Oct 2009
Location: berlin
Posts: 152
Likes: 0
Received 0 Likes on 0 Posts
HN39 I would expect the PF to notice the different 'feel' of a stick with broken spring.
expected yes shure, a broken spring is easy to feel but a worn spring?

did we know how long ago the right SS was not in use? hours? days? weeks?

is it absolut impossible that he feels his (halfback-)stick position as allright?
grity is offline  
Old 25th Nov 2011, 12:18
  #504 (permalink)  
 
Join Date: Jan 2008
Location: LONDON
Posts: 199
Received 21 Likes on 12 Posts
Imagine, if you will, the steering wheel on your automobile randomly varying tire steering angle for a given steering wheel angle. Fun, huh?
Honda, Toyota and BMW have all provided variable gear ratio steering systems.

Electrically Variable Gear Ratio Steering Systems

and they all make some very good cars.

They don't vary anything randomly of course, but then neither does the Airbus.
netstruggler is offline  
Old 25th Nov 2011, 13:10
  #505 (permalink)  
 
Join Date: May 2011
Location: BOQ
Age: 79
Posts: 545
Likes: 0
Received 1 Like on 1 Post
did we know how long ago the right SS was not in use? hours? days? weeks?
Both SS are checked on every taxi out for proper control surface deflection, as well as noting the index cross on the PFD displays for proper SS longitudinal & lateral limits and the neutral position, plus checks of the priority system & dual input advisory.
OK465 is offline  
Old 25th Nov 2011, 13:57
  #506 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Machinbird
Dozy, they didn't know that.
...
It was not an error to deploy the speed brakes when PF did so.
The error was letting PNF cow him into retracting them prematurely.
I understand what you're getting at, but at the same time the PNF seemed to be aware that they had an unstable aircraft - unfortunately we don't (and will never) know if at any point he suspected stall. I'm not an aeronautical engineer, so I don't know what effect throwing out the speedbrakes would have in a mushing stall. I doubt it would be positive.
DozyWannabe is offline  
Old 25th Nov 2011, 14:29
  #507 (permalink)  
 
Join Date: Oct 2009
Location: berlin
Posts: 152
Likes: 0
Received 0 Likes on 0 Posts
OK hours, then the chance is less 1/100.000 that a worn spring plays a role,

(100. 000 are +/- the hours in 10 jears, and in my experience a spring can worn but normaly work longer than 10 jears....)

thats the range of murphys law......
was the SS rescued?
grity is offline  
Old 25th Nov 2011, 16:09
  #508 (permalink)  
 
Join Date: Jul 2009
Location: Not far from a big Lake
Age: 81
Posts: 1,454
Likes: 0
Received 0 Likes on 0 Posts
Worn springs?

Since I presently design products that sometimes use springs, I have some passing familiarity with spring failure modes that may apply to an aircraft feel system. They are:
  • fracture and
  • relaxation.
If you operate the spring into the yield range of stress, you can expect loss of strength over its operating life-This is a gradual degradation.

Otherwise, sufficient cycles can fracture a spring eventually depending on the applied loads. Once a coil type compression spring fractures, the broken ends slide past each other and the spring will shorten, decreasing the effective spring length. This would be the most hazardous failure mode in an aircraft feel system.

In an aircraft feel system such as used on the A330, the springs are enclosed in housings that limit ultimate compression and protect against FOD. The design would avoid stress levels that exceed the Soderberg criterea, thus providing reliable service over the lifetime of the aircraft.

All that you probably ever wanted to know about spring failures can be found here:Mechanical-Spring-Failure-Modes
Machinbird is offline  
Old 25th Nov 2011, 18:22
  #509 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Machinbird
Dozy, please tell me what was bad about deploying speedbrakes?
They were stalled, that's what was bad about it! They were mushing at that point, and slowing down further might have rendered the aircraft completely unrecoverable.
DozyWannabe is offline  
Old 25th Nov 2011, 18:53
  #510 (permalink)  
 
Join Date: May 2011
Location: here
Posts: 131
Likes: 0
Received 0 Likes on 0 Posts
Dozy, what MB is getting at is, rational thinking provided, the marginal effect of speedbrake deployment could have told them that they were very slow. It didn't look much like a methodic attempt at troubleshooting though...
Zorin_75 is offline  
Old 25th Nov 2011, 19:02
  #511 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally posted by HN39 ...

Clever molecules that avoided the ice detectors?
Very clever!

Which leads me to ask if the ice detectors have always detected ice in other recorded UAS events?

Having had another look at the ATSB's report on two UAS events recorded by VH-EBA on 15 Mar and 28 Oct 2009, I failed to find any reference to ice detectors mentioned. Anyway, I extracted the Icing Environment graphic from the report and plotted the AF447/F-GZCP data on it.

The argument for "super-cooled" water being the culprit, goes along the lines that as the probes are already heated, they provide the ideal impurity point for the "seeding" and instantaneous conversion of super-cooled water into ice. Likewise, the melting of this ice occurs in a relatively short time.

Last edited by mm43; 25th Nov 2011 at 19:18.
mm43 is offline  
Old 25th Nov 2011, 20:27
  #512 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Ice detector

mm43;

The ice detector that I'm familiar with (Rosemount, IIRC) detects the formation of ice due to freezing of super-cooled liquid water on its unheated ice-collecting element (*). It does not detect ice crystals floating in the atmosphere. Ice crystals do not adhere to unheated surfaces, but may be 'caught' inside a pitot tube.

(*) Quote from AIAA paper 2006-206 "The Ice Particle Threat to Engines in Flight" by Jeanne G. Mason et al.: In 2002 one large transport aircraft engine powerloss event occurred on an aircraft equipped with dual Rosemount Ice Detectors (RIDs), one on each wing. When exposed to supercooled LWC, ice accretes on the exposed rod of the RID until it reaches a threshold mass, at which point deicing heat is applied to the rod and a ‘trip’ is registered.
HazelNuts39 is offline  
Old 26th Nov 2011, 03:49
  #513 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
lomapaseo;

Re: post #502 (in reply to my post #512!)
please define the boundaries for this relatively short time.
There's that word "relative" again!

In the case of VH-EBA the UAS was for about 8 seconds and the SAT was -48°C. AF447 had UAS for 30 seconds (longer on the ISIS) and the temp was -43°C. VH-EBA had Goodrich pitots and F-GZCP had Thales, so it's not possible to compare directly for any relationship between ice build-up and time, though that's likely.

Neither do we know whether ice crystal accretion over time is responsible, or a near instantaneous build-up due to "super-cooled" water seeding into crystals on contact with the pitots. In the case of both aircraft, pitot blockage/unblockage occurred rapidly.

The "relative" part of your comment can only be answered subjectively on the basis that a small ice accretion will last a short time (seconds) and a larger one will last relatively longer, but still in seconds.
mm43 is offline  
Old 26th Nov 2011, 09:05
  #514 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by HazelNuts39
The ice detector that I'm familiar with (Rosemount, IIRC) detects the formation of ice due to freezing of super-cooled liquid water on its unheated ice-collecting element (*). It does not detect ice crystals floating in the atmosphere. Ice crystals do not adhere to unheated surfaces, but may be 'caught' inside a pitot tube.
As I recall, the issue with super-cooled droplets as opposed to ice is that while contact with impurities cause them to freeze near-instantaneously, the operative term is "near" - they can progress backwards along a moving surface some way in that fraction of a second, and that was part of what caused the problems with the ATR back in the '90s because the droplets ended up freezing as ice behind the de-icing boot. If the droplets don't freze until they are behind the ice detector then is is possible that the ice detectors can't detect their presence in a similar manner?

@Lyman - it's right there in the DFDR traces.



The aircraft climbed due to elevator movement and rising pitch attitude (see red line), *not* because of an updraft.
DozyWannabe is offline  
Old 26th Nov 2011, 09:59
  #515 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
@2:10:08. "I have the controls"
Lyman is offline  
Old 26th Nov 2011, 11:32
  #516 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by DozyW
If the droplets don't freeze until they are behind the ice detector then is is possible that the ice detectors can't detect their presence in a similar manner?
The referenced paper offers a number of arguments for the prevailing opinion that the high altitude power-loss events, and perhaps also the pitot problems, are caused by convective clouds containing a very high density of tiny ice crystals. The point I was making in my post is that the well-known Rosemount Icing Detector does not detect ice particles, since it is designed to detect super-cooled liquid water droplets. BEA says that the presence of liquid water at -40° C is unlikely, and must necessarily have been limited to small quantities. If there was liquid water in sufficient quantity to cause icing problems, I think it is highly improbable that it would have been missed by the ice detector(s). If your scenario was valid, how could they ever detect ice?

I don't think that you are correctly describing the cause of the ATR accident at Roselawn. IIRC, the theory is that the airplane encountered unusually large super-cooled droplets at relatively low altitude. Due to the large droplet size, the area of ice accretion extended beyond the deicing boots. While the ice on the boots is periodically shed by inflating the boots, the ice aft of the boots remained and formed a ridge on the upper surface of the wing just aft of the boots. This is just from memory, feel free to correct me. I don't see any connection to the functioning of an ice detector.

Last edited by HazelNuts39; 27th Nov 2011 at 07:47.
HazelNuts39 is offline  
Old 26th Nov 2011, 13:40
  #517 (permalink)  
 
Join Date: Mar 2002
Location: Florida
Posts: 4,569
Likes: 0
Received 1 Like on 1 Post
mm43

Re post # 514
The "relative" part of your comment can only be answered subjectively on the basis that a small ice accretion will last a short time (seconds) and a larger one will last relatively longer, but still in seconds
OK it's seconds then. Since we are talking about AF447 the time to clear the ice and reestablish valid pitot readings is iimportant.

My feeling is that this time would not be expected to be much different than all the other inflight incidents, in other words seconds
lomapaseo is offline  
Old 26th Nov 2011, 16:23
  #518 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
Dozy (cc Old Carthusian)

Re: Attitude and "climb".

Dozy, The Nose was actually Low by four degrees v/v flight path, so no goalposts were injured in the making of this post. The a/c was "climbing" nose low ........-4 degrees. The auto pilot will trip out with a NOSE Down of 9 degrees, so I offer this theory, try it on.

For thirty seconds prior to loss of Auto Pilot, the VS select shows 5000fpm down.

This value would trip the a/p due excessive Nose Down, so it is rejected by the FCM. Select/Reject for thirty seconds. NO ICE. The a/p eventually latches the excess (+9) value, and trips out at 2:10:05. NO ICE. As the a/p has tripped due to excess control (past limit), the LAW remains in NORMAL. For an additional ten seconds, the a/c is subject to turbulence that duffs the probes, simultaneously. NO ICE. With discrepant normed Speeds, the a/c degrades now into AL2 NO ICE. It is at this point The PNF announces, "We have lost the speeds, Alternate Law..." NO ICE.

For an updraft to lift this a/c (at the recorded rate) is remarkable. It also screws with the AoA vanes. As the Pitch decreases, the vanes read increasingly higher. PITCH at this point is not the a/c's problem, but AoA is. This is why the a/c initially did not "climb" with PF's aft pulled stick. The elevators were trying to catch up with the airstream, for the better part of ten seconds.

At the point that the elevators got back into the "loop" the climb was already breathtaking, and virtually unnoticed in the cockpit.

Were they in the cell? I believe they were......

EG: Can an aircraft climb rapidly with a Pitch displayed as 'Normal'?

Boy Howdy.

Last edited by Lyman; 26th Nov 2011 at 16:44.
Lyman is offline  
Old 26th Nov 2011, 17:45
  #519 (permalink)  
 
Join Date: Oct 2009
Location: berlin
Posts: 152
Likes: 0
Received 0 Likes on 0 Posts
Machinbird, All that you probably ever wanted to know about spring failures can be found here:
thanks fore the link ,

if worn is the wrong word, then sorry I meant relaxation.... is it one spring for pitch or one for push and one for pull into the sidestick??? If I remember well gums has shown a diagram of the energy vs move of the sidestick in thread 3(?) with a kink in the line, so mayby there works more than one spring in each direction

Last edited by grity; 26th Nov 2011 at 17:56.
grity is offline  
Old 26th Nov 2011, 17:55
  #520 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
Read me a little closer, Doze. Whose elevator? Whose VS select?

A/P. Not PF. What are those oscillations in VS (SELECT)?
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.