New hope for search results...
Autonomous Underwater Vehicles may soon find wreckage of Air France Flight 447
not sure if anyone has posted that link before; a quick scan of recent posts and I didn't see any reference to it

Mowing the Lawn
 

Mowing the lawn is an expensive and time consuming process, but what has to be done when you are not quite sure where the lost object has ended up. There was a sonar search for the SS America with towed sonar that took several months of sea time over a few years on a flat seabed

If they succeed in finding AF447 in a few weeks, it will be a sign the technology has greatly evolved.

After reading everything posted here, looking at what information is available at this time, coupled with 40 years of experience of flying;

I am in the 'slow news day non-event' group as well.

Update: "Seabed Worker" & "Anne Candies"
 

The following Radio Navigation Warning to Shipping has been issued by the Brazilian Navy:-

0048/10
NORTHWEST OF ARQUIPELAGO SAO PEDRO E SAO PAULO
CARTA 10(INT. 216)
VESSEL ANNE CANDIES - BLACK HULL AND WHITE SUPERSTRUCTURE AND
VESSEL SEABED WORKER - BLUE HULL AND WHITE SUPERSTRUCTURE
SEARCHED THE WREKAGE AR PERIOD: 01/APR TO 01/MAY
AREA INTERDICTED TO NAVIGATION AMONG PARALELS: 03-48.00N 03-00.00N AND MERIDIANS: 030-22.00W 031-09.00W
CAUTION ADVISED.
CANCEL THIS WARNING 020359 MAY 2010.
------------------
The area to be searched lies within -

3° 48'N 31° 09'W, 3° 48'N 30° 22'W
3° 00'N 31° 09'W, 3° 00'N 30° 22'W

The area covered by this warning is approximately 2,250NM^2, or 7,700km^2, centered on -
3° 24'N 30° 45.5'W.

The BEA have stated on a number of occasions that 2,000km^2 is the area to be searched, which equates to the area of a circle with a radius of 25.2km or 13.6NM.

mm43

mm43,
Using the back of an envelope, and after checking a few different announcements, it seems the search has been narrowed down to about 900 square miles (give or take a few), which is roughly a square of 30x30 miles, which would mean the actual impact point has been narrowed down to 3° 22'N 30° 45.5'W plus or minus 15 miles.

Statute or nautical miles? Not a huge difference anyway...

It still seems a pretty limited area, in view of all the incertainties about the debris field on the surface, and the local currents and winds at the time.
And 600 mph is 60 miles per minute, so again they think they've pinned down the time of impact to within a minute and a half?

It's only just the back of my envelope.... and yes, I truly wish them success, and luck... they'll probably need it!

CJ

ChristiaanJ,

Not disagreeing with your basic point, but my (mental) arithmetic makes the search area:

48 nm North-South X 47 nm (approx) East-West,

which is about 2250 nm²

From memory of flying the route, is the Upper Airway track aligned at about 015 deg(T)?

You're a bit adrift on timing. At a GS of (say) 480 kt, the (flight) time to transit the area would be about 6 minutes.

ChristiaanJ; Chris Scott;

Thanks for alerting me to the error - which has been corrected. I had looked at the warning area and deducted 5NM of each side to allow for a buffer zone, and that's how the original figures came about. http://images.ibsrv.net/ibsrv/res/sr...lies/wink2.gif

The airway track is 028°T, and the Last Known Position was 2° 58.8'N, 30° 35.4'W (3NM west of track), which gives 338.2°T x 27.1NM to 3° 24'N 30° 45.5'W.

Sure hope they get lucky.

mm43

mm43,

Thanks. I forgot to allow for a buffer zone, so search area seems to be more like 38nm X 37nm, and flight transit time on the airway track 028T of the order of 5 minutes.

I assume the BEA have good reason to believe that, unsurprisingly, the aircraft continued to deviate further to the left of the airway centreline?

Only an uneducated guess, but could it be the search area has been reduced because information has come from another source.

i.e. xx Navy submarines which may have picked up low frequency sound from some distance away and have calculated a bearing back to the source.

Chris Scott

The aircraft's Last Known Position appears to have been generated following a LOC incident which occurred with the onset of the various ACARS messages time-stamped initially at 0210z. The AOC position has priority and was injected into the stream at 02:10:30. No comment has been made by the BEA as to the actual timing of the AOC message, but it could be presumed that the ACARS time-stamp was not to the integer of the minute, but was rounded to the nearest minute - or the position was at 02:10:30z.
It is assumed that this deviation to the NNW had happened prior to the AOC position. The radius of the search area is a result of the difficulties presented when trying to backtrack debris drift allowing for current and leeway in a very fickle part of the ocean and at what was the effective center of the doldrums at the time of the accident.

The last satellite contact with the a/c was at 02:14:28 and the flight may have ended at that time or up to 45 secs later, as the BEA have determined that at least one ACARS FLT message remained to be transmitted.

mm43

Search Area and Drift Patterns
 

The following graphic is essentially the same that was posted to the AF447 thread in September 2009. Added to it are the boundaries of the Nav. Warning Area, and a shaded circular area representing the 2,000km^2 the BEA has said they have narrowed the search field down to. A small red triangle which indicates the original calculated backtrack for the V/S using OSCAR/NOAA satellite derived current data along with the QuikSCAT satellite derived 10m wind data, has also been added.

Somewhere back in the original post there is a comment regarding why I increased the current velocity by 50%, which seems now to have been wrong as the best data analysis available to the BEA has concluded it was actually around 50% less.

http://i846.photobucket.com/albums/a...ch-area-1a.png

A larger image is available at - http://i846.photobucket.com/albums/a...h-area-lge.png
mm43

Bank angle
 

I wonder what bank angle a pilot would typically use to change heading to avoid weather, while in cruise at M=0.8, FL350? I don't mean a late, impulsive action or emergency - you look at the radar, possibly discuss with the PNF, then decide to change heading 25 degrees left of track?
regards,
HN39

HN39,

Typically, as you suggest, it would be done in HDG mode, with the AP deciding the bank angle (unless you turned the HDG knob little-by-little). At that altitude, the bank angle would be nearer the 20 than the 35, I think. But perhaps a current A330 pilot will comment.

Chris

10 miles per minute, Christiaan

Rather unlikely
 

60 miles per minute equates to 3,600 miles per hour - would mean a quick trip but is quite unlikely

Chris Scott;
That's exactly how it's done and that's the bank angle achieved for the A330 when using HDG with autopilot ON.

More bank would have to be done manually but I've never seen a diversion done that way.

A bit more detail for those wishing it... A Rate One turn (3deg/sec) at a TAS of 450kts would be about 53deg bank. The maximum bank angle that will be maintained by the autoflight system (autopilot OFF) using the sidestick is 33deg; the maximum bank angle in Normal Law (autopilot OFF) with constant sidestick deflection is 67deg. Positive spiral stability is maintained by the autoflight sysem in Normal Law.

PJ2

mm - to save my digging back to September, can you remind me what the mauve '11-12' etc ponts are?

BOAC

The ORARO - TASIL track mauve numbers are the minute estimates at 463KTS GS based on previous AOC reports. There was a 10KT headwind forecast at FL350, so they represent about 473KTAS. The dots with the white spots in them in the off track deviation are purely speculative timings.

mm43

- yes, it was those to which I referred - whose 'speculation'?

Mine, in the course of trying to graphically represent a scenario where the a/c could have gone based on a backtracked position for the V/S.

mm43

BEA - Press release - 2 April 2010
 

Flight AF 447 on 1st June 2009
A330-203, registered F-GZCP

Information, 2 April 2010

The « Anne Candies » this morning joined the « Seabed Worker », which arrived in the search zone yesterday.

The undersea searches have begun. Further information on these will be released before the end of next week.
-----------------------
EDIT :: Forecast for period from 2010-04-03 1200z to 2010-04-06 0600z is:-
Wind (10m) - 10KT/090T becoming 10KT/060T
Sea/Swell - 1.5m/020T, Period 9 secs becoming 1.5m/060T

mm43

Are there any further updates on the search progress?

There be Dragons Here
 

At the edges of early sea charts, navigators occasionally discovered an ominous warning that "There be Dragons Here." These metaphorical Dragons are just what AF447 encountered in the dark of night last June. Many A/C follow similar courses along the edge of the coffin corner today, so it's critical that this tragic upset be fully understood.

Interest: 2,220,647 views and 4,587 posts -- that was the original AF447 thread archived at the turn of the year. In this newer thread, 132,561 views and already 662 posts.

Sadly, the last post on this thread was "Any news?"

It would seem that the agencies in control and the primary stakeholders -- those paying for the search vessels and the analysis of data -- are uninterested in a pro-active and open information exchange. Bad PR, in the age of T w i t t e r and Blogs. Bad PR, with implications of secrecy and orchestration. Bad PR, when the interest is so high and answers so important.

I'm not sure what activities make level-headed and smart PPRuNers angry, but I should think docile waiting for rare bowls of carefully conditioned information and being shut out of the ongoing big picture would be one of them. These agencies can't say "None of your business" because it is, indeed, your business! None of us wants to fly where there be Dragons.

Note about edit: When this was originally posted, PPRuNe software automatically converted the word "T-w-i-t-t-e-r" to the word "PPRuNe" and the sentence did not read properly... Mods, feel free to correct if the Twi---- word is a proscribed word...

I'm sure those paying for the search vessels would be happy to exchange information, if they had any to exchange.

From what I understand it's still lying four miles down at the bottom of the Atlantic.

The BEA have made it quite clear that they will be releasing further information on the progress of the search by the end of this week. That is a huge advance on what we had in the Phase 1 and 2 searches, and should be welcomed.:D

In the meantime, the weather in the search area is ideal, with a 1 - 1.5m / 10 second swell and 5 - 10 knot winds from the easterly quarter. CAV has no part to play in this operation!http://images.ibsrv.net/ibsrv/res/sr...ies/thumbs.gif

mm43

4 Miles of Patience
 

Thanks GobonaStick for setting me straight. And mm43 for your continuing analysis and excellent briefs!

The BEA website does show increasing availability of information at the press release area along with some videos. Next info promised for later in the week. I suppose a 24/7 live web feed from a camera in the submersibles control room aboard the Anne Candies or Seabed Worker would be too much to ask and boring to most... Be nice to see a daily briefing with some granularity and detail, though. No doubt the entire affair, including the search details, will become more than one best-selling book, and official reports, recommendations, and new rules will be printed in many languages. Let me apologize for sharing my impatience.

GreatBear,
Your frustration is comprehensible, and I'm sure it's shared.

But at this moment, I think GobonaStick's is the only valid answer

Trying to search four miles down in the ocean for the proverbial needle in the haystack is something very slow, difficult and laborious.
And even if you find something, you still have to go down there and make sure it's not another rock....

Do you really want some vague and unsubstantiated information, as happens only too often in the media, and even here on PPRuNe?

Let them get on with the job. They're only too well aware of how many people are interested in what they achieve.

CJ

CJ, your post triggered a thought that may be slightly off topic. I'm putting it out there so that somebody who is talented and in a position to riff on the idea and make it practical might see it.

It is hard to tell one object from another with sonar. It's not impossible. I wonder if it would be possible to build two permanently affixed nodes on the recorder packages that resonate at specific frequencies in a relationship of perhaps 1.7 to 1 in frequency.

I suspect that ratio would hold with pressure. And it would give you, in effect, a nice passive transmitter as part of the boxes. A chirp radar should be able to pick then out of most any other sonar returns that might happen.

If this is a good idea, I hope somebody takes it and runs. I'll shut up now.

{^_^}

BEA - Update: English subtitles to PR Video
 

The BEA have added a link to English subtitles text that matches the commentary on their previously released video of the Recife press conference.

http://www.bea.aero/en/enquetes/flig...es.navires.srt

This now confirms that the search area is between 2,000m^2 and 3,000m^2, and hopefully the next update will give some indication of the actual grid pattern they are using in the search, e.g. is the search being conducted along 350/170T degree lines matching the general line of the ridges and valleys.

mm43

BEA - Update: Departure Video
 

My previous post mentions the English subtitle text available for the PR video. In fact it is actually for the departure video which is available in mp4 format at:-

http://www.bea.aero/fr/enquetes/vol....es.navires.mp4

A Windows Media Player version is at:-

http://countjustonce.com/a330/depart...res_WMV_V8.wmv

This video is informative, and shows the equipment, along with some interesting views of the proposed search area. I will attempt to decipher some of those search area plots and post a graphic with more detail of the search area shortly.

The link in post #670 above, gives the English language text associated with the video.

mm43

QF72 AoA
 

EDIT:: The link to an image of the liftcoefficient-AoA-Mach relation derived from the QF72 upset FDR data has become invalid.
regards,
HN39

BEA - Update on search
 

Flight AF 447 on 1st june 2009
A330-203, registered F-GZCP

Update on sea searches on 7 April 2010

Since the sea searches began on 2 April:

o Orion, a towed sonar on board the « Anne Candies », has covered 600 km2;
o The three Remus autonomous underwater vehicles on board the « Seabed Worker » have made nine dives and covered an area of nearly 800 km2.

The meteorological situation in the zone has allowed the crews to work in good conditions.
-----------------------------

A graphic showing the progress of the search is at:-

http://www.bea.aero/en/enquetes/flig...7042010.en.jpg

This is a high resolution depiction of the search zone.

mm43

QF72 AoA versus cL Graph
 

Thanks for that link, HN39.

Still getting my head round the detail of the graph but, in the light of the discussion some of us had (starting with #523/Mar16th & #537/Mar21st, and concluding here with #558/Mar23rd), it's interesting that they seem to have derived Alpha (AoA) from their stated equation:

" alpha = pitch - fpa "

We have already discussed the possible limitations of this presumption; specifically in rapidly moving air (vertically or horizontally).

I was under the impression that a direct reading of AoA would be recorded elsewhere; either on the DFDR or the QAR, or both; but don't have a list of their parameters to hand. AoA certainly used to be available in real time via the AIDS system on the A320, and it would be strange if that was not recorded and/or transmitted on ACARS. The fact that the QF72 recorders were, presumably, in pristine condition makes one wonder if they will do any better for AF447, when the DFDR is found? The QAR is unlikely to have survived intact.

Who produced the graph, I wonder?

Chris

You are correct Chris, AoA data (2 out of 3) were available in the Interim Report on the A320 XL Airways accident in Perpignan.


http://i65.servimg.com/u/f65/11/75/17/84/aoa_0110.gif

QF72 AoA versus cL graph
 

Chris Scott;
thanks for replying, I’m pleased someone is interested in my little exercise. The equation is correct as long as fpa is defined relative to the air mass. In this case: it is assumed that the airmass does not move vertically: sin(fpa)=vsi/TAS
The DFDR traces in the ATSB report include AoA, but the scale chosen to show the ‘spikes’ is too small to read a real AoA value. Therefore I integrated vertical acceleration to obtain vsi, and from that AoA.
The graph was produced by MS EXCEL from data read manually from the DFDR traces provided in the ATSB 1st Interim Report on the accident, with some help from BOAC and PJ2.
regards,
HN39

HazelNuts39's Graph of AoA versus cL for QF72
 

Quote from HazelNuts39:
"I’m pleased someone is interested in my little exercise.
... with some help from BOAC and PJ2."

You are too modest... I did wonder. But may I suggest it may be advisable to provide an attribution, in case someone assumes it is extracted from an official report?

Quote:"The DFDR traces in the ATSB report include AoA, but the scale chosen to show the ‘spikes’ is too small to read a real AoA value."

I see the problem, but would it still be worth superimposing the official AoA trace on your graph, for comparison purposes: uncorrected – and perhaps also corrected – by the Prandtl-Glauert "rule"?

Not having read the QF72 report in any detail, must admit I do not know if wind and/or vertical air-currents were a factor. In relation to AF447, however, there is every reason to suspect that horizontal and/or vertical gusts may have at least contributed to the aircraft departure.

Quote:
"In this case: it is assumed that the airmass does not move vertically: sin(fpa)=vsi/TAS "

In our earlier discussions, I think I was able to establish in #552/Mar23rd – if only qualitatively – that even a steady movement of the airmass (vertically and/or horizontally) leads to an error in the assumption that " alpha = pitch - fpa ". As for gusts, they are another matter.



Nevertheless, subject to the points above, your graph seems to be a very useful exercise – thanks.

And thank you, CONF iture, for the official graph of AoA for the XL Airways accident.

Chris

QF72 cL vs AoA graph
 

Chris Scott;
Thanks for both suggestions. I've added a note to the graph and the post.
Adding the 'official' AoA doesn't seem practical: The AoA trace is on figure B2, which has a different timescale from figure B4 that I used, and 10 degrees AoA is only 3 pixels.
regards,
HN39

Quote from HN39:
The [QF72] AoA trace is on figure B2, which has a different timescale from figure B4 that I used, and 10 degrees AoA is only 3 pixels.

Sounds virtually useless. I take your point, and presume the investigators would have had access to a better one?

Chris

On that matter, the XL Airways report is not different : The AoA traces are not made available to us in a readable and informative scale ... ?

BEA - Update: Search Operations Video
 

The BEA has released a new video in French showing operations in the search zone, and the mp4 version can be found at:-

http://www.bea.aero/fr/enquetes/vol....tionnelles.mp4

I'll add a link to a Windows Media version as soon as I have converted it to that format.

http://countjustonce.com/a330/first-...onal-dives.zip

Download the zip file, save and then double click on it. The file should then be converted into wmv format and will be ready to play.

mm43