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

AF447

Thread Tools
 
Search this Thread
 
Old 3rd Jul 2009, 04:03
  #2801 (permalink)  
 
Join Date: Jun 2009
Location: LA
Age: 73
Posts: 50
Likes: 0
Received 0 Likes on 0 Posts
Points for consideration

What we can perhaps say that we know with some decent probability at this point is:

- They were almost surely at a known location at 0210 (from the ACARS msg), and almost surely still at FL350.
- They hit icing conditions at about that time that knocked out some combination of the pitots.
- This had various follow-on ramifications in terms of cockpit messages and degraded protections, at the least.
- At some subsequent time they found themselves in the water, within about 30-40nm of their 0210 FL350 position.
- A normal descent from FL350 normally takes more than 30 miles.

We can suppose that:

- There is some indication that they may not have deviated around weather that other pilots that night did deviate around. (The 0210 position does not seem to indicate any deviation, and they were already into icing.)
- They did not deliberately intend to fly in icing conditions
- They may have encountered other weather conditions post-icing event
- They did not intend to land in the water in pieces
- They hit the water in a somewhat horizontal attitude (from the prelimnary report).
- From the 30nm distance made good, we can presume that descent probably began within 10-15 minutes (or sooner) of 0210, else they would have landed somewhere else.

This sequence brings up two questions that seem fruitful:

- What did they hit after hitting icing that caused them to lose control? Is there any answer more likely than another?
- What condition was the plane in as it descended? Stalled? Spinning? Under any sort of control? We certianly can't know, but is there one condition that is more likely than others, if we accept the report's assumption that they hit somewhat flat with some forward motion?

(Edited to change 'conclusion' to assumption' in terms of the BEA report suggesting the plane hit approximately flat.)

Last edited by WhyIsThereAir; 3rd Jul 2009 at 04:14. Reason: add title for introduction of thought process
WhyIsThereAir is offline  
Old 3rd Jul 2009, 04:11
  #2802 (permalink)  
 
Join Date: Feb 2009
Location: The laughing stock of the rest of the world!
Age: 73
Posts: 153
Likes: 0
Received 0 Likes on 0 Posts
WhyIsThereAir...
if we accept the report's conclusion
The report is not a conclusion.
Lightning6 is offline  
Old 3rd Jul 2009, 04:14
  #2803 (permalink)  
 
Join Date: Jun 2009
Location: Colorado
Age: 74
Posts: 48
Likes: 0
Received 0 Likes on 0 Posts
Stall recovering

CC The B-2 is not a conventional aircraft ... I have not flown that type and have no knowledge of its control regime.

For stall recovery in a conventional aircraft you need the tail plane flying first. i.e. elevator authority. If you get the main plane flying first you are going back in the stall ... stalls and flaps don't go well together.

Edit: Failed ditching, stall, spin, VNE+ dive; it's all academic. What caused the upset?

Last edited by EGMA; 3rd Jul 2009 at 04:25.
EGMA is offline  
Old 3rd Jul 2009, 04:21
  #2804 (permalink)  
 
Join Date: Jan 2008
Location: Denver, CO
Posts: 265
Likes: 0
Received 0 Likes on 0 Posts
Piper Driver writes;
Piper_Driver
Yes - you need the engines or APU to maintain pressurization and ... a loss of pitot information as well as dual engine failure ... possible Stall/Spin situation they may have lost engine power. However I still see nothing to suggest engine failure was significant in the accident sequence.
It's unlikely but cannot be ruled out yet - and unlikely as it is, it could turn out to be true.

-- The A-330-xxx has proven itself as a frame which is reliable.
-- Air France has proven itself as a company able to hire, retain, and train excellent pilots.
-- The ACARS transmissions are intriguing but certainly not the "last word" regarding what AF447 experienced. ACARS is not rendundant like flight systems, and the "top" of the aircraft needs to be within 75 degrees of "up" for it to work - so we might not have ALL the transmitted messages regarding flight law, et cetera.

A loss of thrust after a serious upset *would* explain a lot - heavy plane, changing flight laws, degraded instruments...

And, of course, we need to consider all the time required to try to "fix" all the various things gone wrong - huge workload.

Simile:
You CAN clean the rain gutters, replace a sprinkler head or three, mow the lawn, pprune the shrubs, and sweep the driveway -if you plan to on any given Saturday.

But if someone knocks on your door and tells you that you have to get all those things done *immediately* it might make it hard to do any of them, even though you can do all of them "in your sleep," so to speak.

I think one thing we haven't considered in a logical manner in this thread is that the crew might have simply been overwhelmed.

Pilots are never overwhelmed - until they are.

No disrespect intended - but it is true.

I can hand you a list of 10 things to do and tell you to get them done in 10 seconds, and you will.

But should I hand you a list of 15 things to get done immediately, and then turn the lights out, and then start smacking you around, run a strobe light...


The end minutes of AF447 were clearly in the realm of test pilots, not line pilots.

Normally we speak of holes in Swiss cheese - I think here we are speaking of holes in Havarti.


My point is this: Even if we never find out what the actual cause(s) were with AF447, we should take some basic lessons from it.

Many of the respected pilots contributing to this thread have commented negatively on a lack of authorized hand-flying - this needs to change, as there is no substitute for a skilled pilot.

Many have commented on "shi77y" communications along that route. I believe them - and I don't know if this "brownout" exists because the countries won't spend the money needed for more powerful R/T units or whether it is due to weather conditions - but it clearly should be bitched about until it gets fixed.


I'm going to phrase this as "lightly" as I can.

Serious pilots need to hold Airbus accountable for many automation issues. I'm not - repeat, NOT - damning Airbus for their great innovations.

I'm simply saying that YOU should be holding their feet to the fire for every single thing you don't like about the software or the philosophy behind the software.

Personally, I'm in love with the man wearing the stripes in the pointy end, and the man next to him who might need to take over.

And, personally, I'm very hateful of anyone who even considers the possibility that a flight I'm on *should* crash every so many times, based on statistics.
rottenray is offline  
Old 3rd Jul 2009, 04:24
  #2805 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
WhyIsThereAir
- At some subsequent time they found themselves in the water, within about 30-40nm of their 0210 FL350 position.
That inspired me to go looking at the raw reports such as the BEA report showed them. It appears somebody saved 3 bytes by not sending altitude at least as an FL. It surely would be nice to know it's altitude at 02:10:34.

JD-EE
JD-EE is offline  
Old 3rd Jul 2009, 04:29
  #2806 (permalink)  
 
Join Date: Jun 2009
Location: ATL
Age: 67
Posts: 131
Likes: 0
Received 0 Likes on 0 Posts
CC The B-2 is not a conventional aircraft ... I have not flown that type and have no knowledge of its control regime.

For stall recovery in a conventional aircraft you need the tail plane flying first. i.e. elevator authority. If you get the main plane flying first you are going back in the stall ... stalls and flaps don't go well together.
EGMA, not trying to be critical, just relaying my understanding. B-2 was a bad example for the reason you cited, we'll use the before FBW, B-2 flying wing predecessor, and the flying wing Horton? glider. Stalling the tail is bad news and a stalled wing unloads the tail. We do flaps down stalls all the time. Never spun flaps down, and wouldn't want to, but it can be done. With flaps you get the slats on the A330, and wouldn't want to deploy them during out of control unless that was the last option. The Tiger Moth slats pop out in spins if they're unlocked, but that is a very low wing loaded airplane. But the F-4 slats come out during transonic manuevering. The A330 slats weren't designed for spins however.
ClippedCub is offline  
Old 3rd Jul 2009, 04:41
  #2807 (permalink)  
 
Join Date: Jun 2009
Location: Colorado
Age: 74
Posts: 48
Likes: 0
Received 0 Likes on 0 Posts
I agree CC, I've only ever spun clean. I've stalled clean and dirty in a tee tail, I preferred clean; but then I'm an old (not bold) pilot.
EGMA is offline  
Old 3rd Jul 2009, 04:41
  #2808 (permalink)  
 
Join Date: Nov 2006
Location: SoCalif
Posts: 896
Likes: 0
Received 0 Likes on 0 Posts
They would have stalled the A330 in all configurations of gear, flaps and cg in flight test back in 1994. How serious was the damage to the plane in the aft cg stalls?

A full stall series in a DC10, for example, will damage the ailerons and can tear the elevator counterweight horns right off.

For training or refresher, wiki has a fair description of spins and flat spins. The 65 to 90 degrees angle of attack of a flat spin is awfully similar to the BEA report of AF447.

GB
Graybeard is offline  
Old 3rd Jul 2009, 05:07
  #2809 (permalink)  
 
Join Date: Jul 2009
Location: CA
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
Lost in translation?

According to CNN, "They have not found any clothing, he [air accident investigator Alain Bouillard] said, but was unable to say why."

BBC writes: "The French investigation appears to contradict earlier reports attributed to Brazilian pathologists. They said last month that the injuries sustained by the passengers whose bodies had been found suggested the plane had been in pieces before it hit the sea.
Mr Bouillard said France had not yet been given access to those autopsy reports."


The main characteristic that caught my eye with the Brazilian reports was that bodies were found "without any clothing", or "minimal clothing". (no link here, sorry) I'd posted before that it takes more than terminal velocity to strip people of their clothes - witness sky divers. I had concluded that the only way bodies could get stripped off their clothing is by getting ejected into air traveling at much higher speeds, e.g. Mach .8 or such. Which suggested that the plane did indeed break up at high speed and high altitude(?).


However, high speed and low altitude (ocean level) breakup of an AC could have the same result. This would require bodies getting ejected upward from the AC even as they had traveled downward (vertical component) at a high speed. Only some sort of rebound effect could explain that.


Or, is the phrase "without clothing" meant just like CNN reported, that there was no clothing (apart from bodies) found. Which, however, ALSO seems ambiguous. Does anyone have any better insight into this?



The BEA report does sound very convincing - I am trying to figure out how to account for other evidence within the context of that report. It sure would be helpful if the relevant agencies could get their acts together.
Mike HD is offline  
Old 3rd Jul 2009, 05:35
  #2810 (permalink)  
 
Join Date: Jul 2009
Location: CA
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
ttcse

"Perhaps many wet fabrics sink" - they do, and tend to drag you down. As I learned with training as a water safety instructor, needing to swim several lengths of an olympic pool with shoes, socks, jeans, and a sweatshirt. They REALLY drag you down. Which, BTW, removes you from the "wave action" you adduce:

" I imagine the attire of passengers traveling from Rio would be no challenge for waves to remove." Those waves must have very nimble fingers to unbuckle belts and unbutton shirts. And, perhaps undo shoelaces, too?

I didn't intend my question to be taken frivolously - if BBC or any such news agency perks their ears on something that had caught my eye - well, maybe there is just a tad of credibility to that?
Mike HD is offline  
Old 3rd Jul 2009, 06:21
  #2811 (permalink)  
 
Join Date: Nov 2000
Location: not a million miles from old BKK
Posts: 494
Likes: 0
Received 0 Likes on 0 Posts
I'm really confused now. From the London Telegraph:
Investigators said the Airbus "descended vertically" and dropped 35,000ft in a matter of seconds, hitting the water in its exact flying position.
Now, for an object to fall around 7 miles in 60 seconds it needs a terminal velocity in excess of 400 mph. The inferance that 'it fell in a matter of seconds' is that it completed the distance in less than 60 seconds putting it at or near supersonic speeds. And it pancaked, intact after all that?
I know that Mother Nature has her moments but I have never heard of a downdraft so powerful as to have accomplished this and if the aircraft hit the water 'in its exact flying position' then it certainly wasn't an engine assisted dive or do they mean it's position at 35,000 when the catastrophe struck?
What are the French trying to tell us here I wonder?
Xeque is offline  
Old 3rd Jul 2009, 06:34
  #2812 (permalink)  
PJ2
 
Join Date: Mar 2003
Location: BC
Age: 76
Posts: 2,484
Received 0 Likes on 0 Posts
rottenray;
And, personally, I'm very hateful of anyone who even considers the possibility that a flight I'm on *should* crash every so many times, based on statistics.
You're not whistling Dixie.

This is not the thread to take further the excellent points you have raised and with which many cockpit crew will agree including myself, (now retired).

The points you raise have been discussed elsewhere at length and deserve greater exposure; this is not the first time they have been raised and should form the lessons this industry, including cockpit crews and their representatives, should take from this, and other recent fatal accidents.

Previous posts in this thread have made it clear that for those who fly the aircraft and for those who fly commercially, period, this is a serious discussion on fundamentals and not an uninformed, anti-Airbus rant. What needs saying applies equally to Boeings.

Nor does this focus on the crew of AF447. Few if any here would not be similarly challenged.

Well said, sir.

PJ2
PJ2 is offline  
Old 3rd Jul 2009, 06:44
  #2813 (permalink)  
 
Join Date: Jun 2009
Location: Colorado
Age: 74
Posts: 48
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Xeque
I'm really confused now. From the London Telegraph:

Now, for an object to fall around 7 miles in 60 seconds it needs a terminal velocity in excess of 400 mph. The inferance that 'it fell in a matter of seconds' is that it completed the distance in less than 60 seconds putting it at or near supersonic speeds. And it pancaked, intact after all that?
I know that Mother Nature has her moments but I have never heard of a downdraft so powerful as to have accomplished this and if the aircraft hit the water 'in its exact flying position' then it certainly wasn't an engine assisted dive or do they mean it's position at 35,000 when the catastrophe struck?
What are the French trying to tell us here I wonder?
I think you'll find that is a media assumption ... if the SLF didn't have their life jackets on it must have been quick. Nowhere in the report does it mention a descent rate.
EGMA is offline  
Old 3rd Jul 2009, 07:27
  #2814 (permalink)  
 
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes on 0 Posts
For those of you who have been with us for some time, here's some of what the report says that's interesting, or novel.

ACARS messages:

Anyone interested in the problems of sorting out the timing of ACARS messages should read the report, as it lays in out detail the logic behind the transmission order, and the shortcomings of using ACARS messages in this way. Faults are accumulated and transmitted in hierarchical order every minute. Warnings are transmitted in real time. Not every problem results in a fault, and not every warning corresponds to one either.

They do not have an explanation for the TCAS fault at this time.

The do have reason to suspect that the 30-second+ intervals in transmissions starting at 0213 are caused by an interruption of the satellite connection.

There's an automated position report at 0210. (this has been noted already here)

Weather:
at 0210, AF447 was in the middle of a mass composed of 4 CB cells that had grown together (an hour and a half before?), with tops around 52,000 feet.

Weather information:
There's one little note: the SIGMET that expired at 0200 may have contained an error:

WSBZ31 SBRE 311752
SBRE SIGMET 5 VALID 311800/312200 SBRE-RECIFE FIR EMBD TS FCST SW OF CLARK PSN/ PEPER PSN/ NEURA PSN AREA TOP FL350 STNR NC=
On peut s’interroger sur la validité de ce message, pour deux raisons :
ˆˆ la ligne AB est orientée sud-ouest / nord-est or, en pareil cas, on attendrait une mention SE OF ou NE NW OF, et non SW OF comme c’est le cas dans le SIGMET,
ˆˆ les images satellite pendant la durée de validité du SIGMET positionnent
la zone de convection plutôt au nord-ouest de la ligne AB qu’au sud-est.
Il est donc envisageable que ce SIGMET comporte une erreur, avec une
mention SW OF au lieu d’une mention NW OF. Dans cette hypothèse, le SIGMET 5 SBRE viendrait compléter la zone identifiée dans la FIR ATLANTICO, comme représenté par le tracé vert dans la figure ci-après
(I'm assuming for 'sud-est' on the second bullet, they meant 'sud-ouest')

Radar:
This has already been treated above. Other flights saw the weather on radar (when placed in the correct mode). AF has a detailed chapter on wx radar operation in their "supplementary aeronautical manual" which "is not imposed by regulation".


In short, we know that they flew into the middle of a huge thunderstorm without any meaningful deviation. The flight went down sometime thereafter. The interruptions in ACARS messages imply that an upset may have occurred as early as 0213.
DingerX is offline  
Old 3rd Jul 2009, 07:38
  #2815 (permalink)  
 
Join Date: Aug 2002
Location: Switzerland, Singapore
Posts: 1,309
Received 0 Likes on 0 Posts
intact on impact

I think this thread goes lost even more after the first report from Bea.

Let me clarify the thing about "intact on impact":

Bea says that the A330 was "not destroyed in flight". They can tell by the damage scars on the debries they found.

Now, what does that mean exactly? It doesn't necessarily mean that the aircraft flew (controlled or uncontrolled) into the water. It merely means what it says: it was still one entity at the impact.

It's very unlikely that they ditched. It's very unlikely that they glided towards the water for any reason.

The most likely theory is still the one with the iced probes, the unreliable speed and the inaptitude to control it. They got some stalls and some overspeeds. They may have lost some critical control surfaces (remember: they only found the tail, but not the elevator or the wings), the aircraft got uncontrollable and fell down, everything between high and low speed.

This aircraft was still "complete" as a whole. But it was not controllable. That's what Bea wants to tell us. Nothing else, nothing more.

Dani
Dani is offline  
Old 3rd Jul 2009, 07:40
  #2816 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
On the topic of 'translation', I have only the English report. Can anyone tell us (from the report) how the BEA appear to have deduced the 'heading' at impact (?if what I read here is correct!?) and how we all appear to have 'accepted' a near vertical descent rather than some sort of varying heading during descent - or even 'turn-back' - resulting in impact near the last known position?
BOAC is offline  
Old 3rd Jul 2009, 07:47
  #2817 (permalink)  
 
Join Date: Aug 2008
Location: London, UK
Age: 68
Posts: 37
Likes: 0
Received 0 Likes on 0 Posts
jeremiahrex



Quote:
While it is indeed possible to extend radar coverage to every part of the globe...
Not with ground/land based radar it isn't. Radar is line of sight and the earth is curved. Even at altitude, an aircraft will often be below the horizon WRT to the nearest land.
TiiberiusKirk is offline  
Old 3rd Jul 2009, 08:03
  #2818 (permalink)  
 
Join Date: Jun 2009
Location: CPT
Age: 57
Posts: 6
Likes: 0
Received 0 Likes on 0 Posts
Leandro:

If as the BEA report suggest .. the plane don't broke in air before touch the water .. why no more ACARS messages durring the fall
No more defaults to be detected and send ... ?
Non-frequent SLF here (CPT to Europe maybe once/year), fascinated by the physics of flight; and I have learnt a mass of things about said physics and flight control systems by following this thread.

Some tentative questions: Is there a consensus that the ACARS messages were transmitted prior to the aircraft exiting normal flight? Is it possible that the ACARS transmissions occurred after that exit? In other words, things were already starting to go wrong in terms of normal flight and the ACARS were more or less transmitted at the same time, or that there was an overlap between transmission and sub-normal flight. Extending this further: the end of ACARS transmissions doesn't necessarily point to break-up at cruise level?

Thanks.
Rustum is offline  
Old 3rd Jul 2009, 08:10
  #2819 (permalink)  
 
Join Date: Aug 2002
Location: Switzerland, Singapore
Posts: 1,309
Received 0 Likes on 0 Posts
Aircraft with unbalanced control surfaces (i.e. if you loose some of them), tend to oscillate between the extremes (speed and attitude). So they go nose up, speed decrease, stall, nose down, speed increase, overspeed. And so on. Until it hits the surface. How it hit we still don't know. But obviously just in a straight level, which is pure randomly, it could also have impacted nose down. It all depens in which phase of the oscillation it was.

Concerning the VS I don't know what you're refering to.

As I (and others) stated long time ago: The ACARS messages don't deliver the cause of the accident, but the outcome. As soon as the aircraft spirals down, you get these weird messages. So it's very well possible that they were created after the aircraft became uncontrollable.

Dani
Dani is offline  
Old 3rd Jul 2009, 08:18
  #2820 (permalink)  
 
Join Date: Jul 2009
Location: Bristol
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
Hi all,

I've been lurking on here for quite a while but I though I might post on a though I had after reading all the posts on here and the report yesterday.

First off I'd just like to say I am not a pilot and I dont work in civil aviation. However, I am an engineer with 10 years experience and an aerospace background. Several years back I took part in 2 crash investigations regarding unmanned military aircraft.

The indications seem to be that the aircraft went into the sea at a high rate of speed but also it would seem it lost considerable altitude in a very short time. This does not seem consistent with the fact the aircraft met the ground in one piece - how could so much altitude be lost so quickly?

My suggestion is that the pitot tubes could have failed (iced up) quite some time before the accident or any indication of a fault. Suppose all three tubes failed at approximately the same time holding the readings for cruising altitude and speed. The consistency in readings may have got past the computer which may not have raised the fault.

In this eventuality is it possible that the auto pilot could have continued to operate on false information leading to gradual loss in altitude over a relatively long period and hence this was not noticed by the crew or ATC which would get altitude from the aircraft via secondary radar.

The pitot fault messages could have been caused as the pitot tubes began to de-ice as a result of the warmer air at lower altitude. If they did not de-ice at the same time this would cause the disagreement which initiated the fault.

After that the crew would have had a high workload and very poor situational awareness due to the instrumentation fault. Perhaps the presence of weather also blocked out the moonlight making any visual cues difficult to see.

Could the crew have flown into the sea at high speed if they had much less altitude than expected? Maybe the tail first impact suggests the only saw the sea at last minute and pulled up resulting in the tail striking the water.
thescouselander is offline  


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.