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

AF 447 Search to resume

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

AF 447 Search to resume

Thread Tools
 
Search this Thread
 
Old 2nd Dec 2010, 13:28
  #2541 (permalink)  
 
Join Date: Jan 2002
Location: USA
Posts: 286
Likes: 0
Received 0 Likes on 0 Posts
From the beginning of this thread I have felt that the event occurred very quickly, and a rapid deterioration of flight occurred. The suspected fact that paxs were without life vest donned and that cabin crew were not at assigned, or any station seems to validate this. Anybody not seated, or probably belted, at the time most likely never reached a seat and were at the mercy of where they were placed at the onset of the event.

A stall would have ultimately placed the airplane on any random heading, particularly with the drop of the right of left wing. Resulting spin and rate of descent could extend the path in any direction. However, drift estimates should not be discounted. Can one say it is really the only simi hard fact to hang your hat on.
wes_wall is offline  
Old 2nd Dec 2010, 13:44
  #2542 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,224
Received 412 Likes on 257 Posts
ccp6f
While I appreciate your effort at outlining some base scenarios, you mixed in a bit of hyperbole with a premise and conclusion there.
(c) The autopilot is unstable (in severe turbulence) with loss of airspeed data and caused the plane to stall. Had the pilots been in direct control of the aircraft the entire time, this would not have happened. This is corroborated by the incident on Quantas flight 72. Thus the flight control system is poorly designed and should be made more robust or the FBW idea should be scrapped altogether. This would be catastrophic for Airbus.
Your bit in italics seems to me a false dichotomy. FBW is a design practice that is used (successfully) throughout much of aviation - see for example high performance Military Jets since the F-16 or so.

If the Flight Control System in this, and or other Airbus designs (by which I assume you mean all other than the flight control surfaces themselves) could use an improvement, revision, or correction, why do you bring up "scrapped altogether" as an OR statement? That's a non-option, not only due to its risk to Airbus and its business base. Continually tweaking and correcting this, that, or the other is how the aviation business has been operating for some decades. Small improvements here and there, sometimes triggered by tragic events such as AF 447 ...

FBW is an industry standard that works well enough, regardless of my own or others' discomfort with some of its drawbacks.

From the early pages in this thread, a non-trivial amount of discussion has addressed the point that points to pilots having to fly/act when the auto functions degrade or change mode, for one reason or another. That doesn't mean that FBW needs to be scrapped, but it does require any professional aviator to know his aircraft systems inside and out, and to be familiar with degraded modes for any aircraft system or sub system.

Any pilot is going to be challenged if, quite suddenly, airspeed indication becomes unreliable. It's a key piece of information that is the foundation of any flying decision or action a pilot undertakes.
Lonewolf_50 is offline  
Old 2nd Dec 2010, 15:10
  #2543 (permalink)  
 
Join Date: Dec 2010
Location: US
Age: 42
Posts: 5
Likes: 0
Received 0 Likes on 0 Posts
Thanks for pointing that out Lonewolf. I think I confused terminology and probably shouldn't have said that part. I meant the post to be a question, but I think it came out as more of a statement of fact. My aviation experience is limited to flying hang gliders, a lot of reading, and riding in the cockpit of my brother in law's 737 (who flies for a 3rd world airline where regulations are more like suggestions than law) so I really don't have much expertise in any of this. This crash has just been haunting and puzzling me ever since it happened and the engineer in me really wants to understand what might have happened. What I was trying to ask is what this accident could mean for the Boeing vs. Airbus flight control philosophy if it turns out that the computer stalled the aircraft. Anyway, I'll edit my previous post to reflect better what I was trying to say.
cpp6f is offline  
Old 2nd Dec 2010, 16:52
  #2544 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
slats11

Back in August 2009, I had a look for available wind data over the first 6 days of June, and put together a small montage showing the Quikscat satellite 10m data. Its reproduced again below.



Like takata, I came up with some different scenarios as to where the aircraft could have gone and why. Some were based on researching available current and wind data, and they placed the aircraft to the west of LKP. Others used LOC stall and spin scenarios and placed the aircraft relatively close (20NM) to LKP and to the east, but the current setting to the NW was not verifiable, though it was the seasonal direction published in some authortive texts. However, I've since learned how fickle the surface currents are in this area, and any of the positions I developed could be correct.

The one 20NM to the ENE of LKP (3°05'N 30°18'W) has not been covered by side-scan searching.

mm43

Last edited by mm43; 2nd Dec 2010 at 17:52. Reason: added seasonal direction
mm43 is offline  
Old 2nd Dec 2010, 22:33
  #2545 (permalink)  
 
Join Date: Aug 2007
Location: sydney
Age: 60
Posts: 496
Likes: 0
Received 2 Likes on 2 Posts
Hi mm43

As far as bodies are concerned, you can largely ignore the wind. This will have minimal effect on a mostly submerged human body. The overwhelming majority of body drift will be due to currents rather than wind. The wind component will be far smaller than the uncertainty about the velocity of the current, and so it is probably easiest to simply discount the wind this as a factor when attempting body drift analysis.

If you wished to allow some wind component, the winds were easterly rather than westerly. This would also suggest that east is the direction to search. The currents were pretty much due north for 5 days after June 6. Who knows what they were doing prior to this, but they were likely northerly rather than southerly. Hence SE is the best option - which also allows for the pollution spot if you chose to incorporate this in any explanation.

The easterly winds explain why so much other debris (which will have been much more wind effected) was found west of the bodies.

If you had some homogeneous debris (eg 20 seat cushions), then you could attempt some composite analysis. Assume the bodies were 100% current effected and the debris was some composite of wind and current, and then look for solutions. Unfortunately the debris is all heterogeneous, and hence each has its own ratio of wind and current effect. Thus the debris is really just noise, and it is best to confine the analysis to bodies (the only things which will have behaved similarly).

The bodies are all very tightly grouped together. Remarkably so after 10 days in the open sea. We know the current was pretty constant after the first body was discovered. You could speculate that the current had not changed direction a great deal prior to this - on the basis that changes in currents will favor dispersion and scatter. It probably changed speed however (slower prior to discovery of the first body) - because there is no logical explanation for the plane to have got to Takata's original position. This however is all very qualitative rather than quantitative in nature.
slats11 is offline  
Old 2nd Dec 2010, 22:55
  #2546 (permalink)  
 
Join Date: Sep 2009
Location: Around the World
Age: 74
Posts: 87
Likes: 0
Received 0 Likes on 0 Posts
Current and Drift ...

mm43

Ref: your previous messages #2479 and #2491

Since a few days, I can't understand what is the real North equatorial current! (sure, it is very changing twice a year).

I had been very surprised reading pages 30 and 32 of this report (released 2010,oct. 5).

If pages 28 or 30 are very well known , the nine drifting buoys tracked in page 32 clearly show that south-EST zone investigations might be very interesting (Unfortunatly, I don't know what day/mounth/year Coriolis data base is refering).

Regards

Last edited by NeoFit; 2nd Dec 2010 at 23:03. Reason: modify wrong "South West" for "South Est", of course.,
NeoFit is offline  
Old 3rd Dec 2010, 01:05
  #2547 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
NeoFit

The eight Self Locating Data Marker Buoys were air-dropped between 0330 and 0430z on 04 June 2010. They were tracked for many days, and perhaps you and Slats11 will now have a better appreciation of why I was trying to explain that the currents within this region are as unpredictable as a chimp's handwriting.



The Orange and Yellow drifters dropped to the east of the LKP exhibit the greatest commonality in their tracks. They have effectively drawn what is known as the Equatorial Counter Current, while the others have fallen into "no mans land" and have made various rates of progress, in a generally clockwise direction.

Are these tracks in anyway representative of what happened in the first week of June 2009? No they are not, and I would offer that the pattern observed in the graphic above will never repeat again.

The point I have made before and shall reinforce again, is that the matching of what happens in practice with the events that cause it, is not simple. Much effort has been made to date to try and replicate these tracks using many adjusted variables. Until the methodology can be refined, and a better understanding of the small physical changes that propel the surface currents in this area, our ability to backtrack with any degree of confidence is quite limited.

Slats11

There was low profile debris found to the east of the general line in which the bodies were found. I am not going to argue the point over the leeway issue, except to say that they did make a small amount of leeway for reasons I explained earlier. The 10m wind (with lower level surface friction taken into account) was used in calculations I made with regard to the leeway made by the Vertical Stabilizer, and can be found in the old AF447 thread.

Now you could go and seek out the NOAA-OSCAR surface current data for 5 and 10 June 2010 and see how it compares with the drifters above. You'll be surprised.

mm43
mm43 is offline  
Old 3rd Dec 2010, 02:21
  #2548 (permalink)  
 
Join Date: Aug 2007
Location: sydney
Age: 60
Posts: 496
Likes: 0
Received 2 Likes on 2 Posts
mm43

I think we are furiously agreeing that the currents in this area are highly variable, and that any backtracking calculation will inevitably carry a large amount of uncertainty. Certainly the actual observed currents bear little relationship to the low resolution charts showing a NE current. I am not surprised about the disparity here - I have repeatedly tried to make the point that these charts can't be used.

What we do have is:
a) 8 boys dropped in a box around the LKP on June 4
b) in the order of 100km drift west as soon as the buoys were dropped. At the likely speed of the current (and the tracks are not time stamped), this may represent 2 days drift to the west.
c) many of the buoys then turned north - probably around June 6.
d) a fair way north of here, we then had bodies tracking pretty much due north from June 6 to June 10.

What happened before June 4? Who knows. The NOAA charts sure as hell can't tell us.

Do you have time stamped data for these charts? Does the turn to the north (from this graph, many of the buoys did turn north after heading west) fit with the time when we know the bodies were drifting north? That would be interesting to know. If the buoys and the bodies were drifting north at the same time, then it is probable that they were drifting generally west at the same time.

Anyway, the total OBSERVED drift of the bodies was to the north and likely also to the west.

It would be fairly unlikely (although of course certainly not impossible) that the drift in the first 4 days was in an opposite direction and of sufficient magnitude to totally offset this subsequent drift. Hence it is fairly likely that the total drift (OBSERVED and NON-OBSERVED) was to the NW. And hence the impact point is somewhere to the SE. How far away is anyones guess. I can't state it any more clearly than that.
slats11 is offline  
Old 3rd Dec 2010, 02:50
  #2549 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Slats11

You missed the point. The buoy drop was a year and 3 days after the accident which happened on 01 June 2009. The only relevancy is the time of the year, but that's as far as it goes.

mm43
mm43 is offline  
Old 3rd Dec 2010, 03:10
  #2550 (permalink)  
 
Join Date: Aug 2007
Location: sydney
Age: 60
Posts: 496
Likes: 0
Received 2 Likes on 2 Posts
Sorry - I did miss that. I assumed you were presenting data from the time of the accident. So ignore much of what I said.

What datum buoy information from the time (ie during the SAR phase) has been made publicly available?
slats11 is offline  
Old 3rd Dec 2010, 04:21
  #2551 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Very little. I've seen some, but they're as relevant as the lot above.

The North Brazil Counter Current provides the NW push into the LKP 40NM radius circle, and at some point the North Atlantic Equatorial Counter Current forces a turn to the North. The Brazilian current does that (turns north) and around 4N backs to the west, while the Counter Current veers to the east. The whole thing is extremely fluid (no pun intended!) and the if, where, when and how are very much a lottery.

Its the complex geophysical forces in the equatorial region that are stretching the known boundaries. Backtracking in theory should be easier than forecasting, as you have the immediate past environmental history to work with. In this case its knowing how to apply the history that's the killer.

mm43
mm43 is offline  
Old 3rd Dec 2010, 10:55
  #2552 (permalink)  
 
Join Date: Aug 2007
Location: sydney
Age: 60
Posts: 496
Likes: 0
Received 2 Likes on 2 Posts
I agree that is the general pattern. The obvious problem is that there is no way of knowing exactly where these currents change - in either direction or magnitude.

However I think we agree that the bodies likely drifted NE or N or NW for most of the time before they were found. This means the impact point is a fair way south of where the bodies were first found. I don't think we can be more specific than this.

So I wonder why the search area has been so far north - north even of the first bodies found. On the basis of drift analysis alone, this is hard to understand. Presumably those responsible have access to other information that points elsewhere. And I am sure that all the information available is somewhat contradictory and doesn't all point to the same spot. It would however be interesting to know what this other information was.
slats11 is offline  
Old 3rd Dec 2010, 11:32
  #2553 (permalink)  
 
Join Date: Mar 2001
Location: us
Posts: 694
Likes: 0
Received 0 Likes on 0 Posts
slats11, on pdf p. 18 (of 35) of the BEA document NeoFit linked to, (thanks for the link) the blue line shows quite precisely the cumulative boundary of the aerial searches flown by Brazil and France through June 5, 2009 in the area of the last known position. Again, if large pieces of the wreckage, e.g., the VS, were in this area prior to drifting north or west and being located and recovered, they were missed.

Thus for your theory based on takata's original hypothesis -- while I can't speak for him and I don't believe he has posted on this thread in many months, even he backed off his hypothesis if I recall correctly -- to be credible, the Brazilian and French search was very inadequate, if not incompetent, to have missed such large objects.
SaturnV is offline  
Old 3rd Dec 2010, 12:18
  #2554 (permalink)  
 
Join Date: Aug 2007
Location: sydney
Age: 60
Posts: 496
Likes: 0
Received 2 Likes on 2 Posts
I am sure they did the best they could, but I think we have to be careful assuming the aerial search would have found whatever was there to be seen. The effectiveness of the search would have been compromised by many factors.

We know the bodies were well within this blue line as of June 5. That is beyond doubt. And they were not found until June 6. So they missed 50-odd bodies. What else did they miss?

We must be very careful not to close off any possibilities prematurely. I think there is a real risk of doing so if we make assumptions based on what was not see where during the aerial search phase.

The aerial search may well have ultimately been more useful if they had dropped more datum buoys as soon as possible. Then drift analysis would be based on real data rather than conjecture. Of course it was a SAR mission at that time, and the focus was understandably looking for possible survivors. Not on obtaining current data.
slats11 is offline  
Old 3rd Dec 2010, 12:34
  #2555 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,333
Received 104 Likes on 51 Posts
The bodies are all very tightly grouped together. Remarkably so after 10 days in the open sea
That was also my impression.
The drift of the bodies represented a rather sharply shaped Funnel.

What I don't understand:
Why wasn't the First Search performed in an opposite direction funnel shape of let's say 20 -30° either side beginning around the first body found going opposite direction of the center of the body drift funnel i.e. pretty much south from there.
From my point of view it is not extremely probable that the drift direction during the first six days was in a completely different direction and the bodies still being rather close to each other after June 6th.
You see it with the buoys.
When the direction changes dramatically the drifting objects won't end up close together.
The bodies being relatively close together after 10 days points to a rather constant drift for me.
Yes that would not be a really systematic box search, but if you lost some personal papers and found some in the shape the bodies were found, how would you search for them ?

regards,
henra

Last edited by henra; 3rd Dec 2010 at 13:34.
henra is offline  
Old 3rd Dec 2010, 13:19
  #2556 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
Likes: 0
Received 0 Likes on 0 Posts
Morbid...please refocus!

This thread is MORBID. Why not exercise your minds as to the timely intervention of both the aircraft maker and the authorities as to the issues with the pitot equipment.
DERG is offline  
Old 3rd Dec 2010, 13:48
  #2557 (permalink)  
 
Join Date: May 2010
Location: PLanet Earth
Posts: 1,333
Received 104 Likes on 51 Posts
Why not exercise your minds as to the timely intervention of both the aircraft maker and the authorities as to the issues with the pitot equipment.
So that means you basically accept that if the pitots on an airliner fail it is OK for it to drop out of the sky ????
That is not seriously your take on this accident ?!

It has to be found out why after the assumed failure of the pitots it ended up where it did and what can be done in future to prevent this - be it by design improvements or training or whatever comes out of it should the boxes ever be found.

There will always be the possibility that the pitots will freeze. We will probably not be able to totally convince mother nature of not ever doing that again to any airliner.
henra is offline  
Old 3rd Dec 2010, 13:50
  #2558 (permalink)  
 
Join Date: Mar 2001
Location: us
Posts: 694
Likes: 0
Received 0 Likes on 0 Posts
The backward drift analysis graphic is on pdf p. 24 of the BEA PPT presentation.

http://www.asso-af447.fr/images/docs...bre%202010.pdf

I believe the ACARS triangle marks the LNP.

On the graphic on p. 18, I believe the dot marked MF2 is the Meteo France calculation of the most likely impact point. As can be seen, it is to the left of the leftmost boundary of the search grids flown through June 5. I believe the grid that came closest to this point was the first day's search.

The Meteo France estimate of the impact point (at least that's what I assume it represents) is more clearly displayed on p. 14.
________________________

As to why initial search grids focused on the right of the track, I've not seen an explanation. Perhaps, the Mesoscale Convective System was drifting west to east, and if there had been an in-flight breakup, winds would have carried parts of the plane to the east. Perhaps, ATLANTICO noted that AF 459 following had ultimately deviated to the right when encountering the same MCS, and thought AF 447 might have done similar. When debris was spotted on the second and third days well to the south and east of the track and LNP, search efforts were concentrated there until the sightings proved to be false.
SaturnV is offline  
Old 3rd Dec 2010, 15:56
  #2559 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
Likes: 0
Received 0 Likes on 0 Posts
Pragmatism

They have not got a hope in hell of finding the pitot tube.

They do have a hope in hell of avoiding the issue that faults were know and not addressed promptly.

Some of you guys are obsessed with bodies and body parts. refocus you minds on the pitot issues.

All this serach will do will prolong the litigation by Air France and the relatives through the courts. I find this morbid discussion offensive.
DERG is offline  
Old 3rd Dec 2010, 16:28
  #2560 (permalink)  
 
Join Date: Jan 2002
Location: USA
Posts: 286
Likes: 0
Received 0 Likes on 0 Posts
Well DERG, if the bodies can be tracked back to where they came from, then it follows that the pitos will be in close proximicy. Since the subject of this thread is the search for AF447, I think the recent posting(s) has been right on point.
wes_wall 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.