AF 447 Search to resume (part2)

[sensor_validation]

Originally Posted by HazelNuts39

...
I am slightly disappointed by Appendix B. The timescale is unreadable and, while the story is about Mach and AoA, there is no trace of either. There is a trace of a 'MMO Exced.' discrete, and I suspect the DFDR also contains discretes for alpha-prot and High Speed protection, which are not shown.

If you compare the 'Pitch' trace and 'MMO Exced.', the High Speed protection may have been triggered briefly, but the pitch trace really takes off after the Mmo exceedances at 14:21:50 hrs, ten seconds after AP disengage.

Regards,
HN39

I think we can safely assume the grid marks are 60,10,5 seconds. - check the RA lasted 27 seconds, and max rate of climb 6000ft/min.

The elevator action presumably triggering the pitch response clearly takes place only a few seconds after AP disconnect, and before AT off, apparently coincident with the second MMO exceed event.

I am intrigued by the comment re alpha-floor - responsible for the N1 max while AT off?

The altitude seems very heavily filtered/ phase delayed - maybe to do with fact that it is "Pressure Altitude"?

[grity]

HazelNuts39 I am slightly disappointed by Appendix B. The timescale is unreadable and, while the story is about Mach and AoA, there is no trace of either. There is a trace of a 'MMO Exced.' discrete, and I suspect the DFDR also contains discretes for alpha-prot and High Speed protection, which are not shown.

the time-lines are minutes......so this zoom climb excursion was 100 s long until the A340 again reach his old flight level

but what is the definition from PITCH while flying through a turbulenz, angle to what???

the pitch was only +1 deg in the moment the "AoA law (?)" let move the elevator to +4 deg for abaut 2-3s and start the zoom-climb, just an overreaction of this system?

and for me it looks as if the "AoA law" did not stopp with the first forward-sidestick-input, this was 15 s more or less forward, and the elevator moves happy up and down and up and down....in the range +2.......-2

[HazelNuts39]

sensor_validation;

Consider these extracts from FCOM 1.27.20 p.5, High Speed Protection:
Quote: When it is activated, the pitch trim is frozen. (...) The High Speed Protection is deactivated when the aircraft speed decreases below VMO/MMO, where the usual normal laws are recovered. Unquote.

Furthermore, in normal law with AP off and no sidestick input, the FCS maintains 1g. The elevator movements you noticed may be the system's normal response to g-variations due to turbulence.

Regards,
HN39

[Centrosphere]

slats11,

discussions about shark bites are devoid of significance and potentially distressing to "interested parties."

They´re not so "devoid of significance" because this can be of interest in assessing the amount of damage inflicted to the bodies that were recovered floating, and to evaluate how useful they could be in the investigation concerning what happened to the plane.

Of course, if everything else fails (the black boxes refuse to throw out their secrets, etc), every bit of evidence will be useful. In this case any alternative should be persecuted, like try to recover the bodies at the bottom, or filming it in HD 3D, or something else.

Sadly, I´m used enough to internet foruns to know that at this height anything I write here under the name "Centrosphere" will be imediately answered with hostility by some people, since it seems I inadvertently made some enemies here. But consider this:

Even the posts concerning the nuts and bolts of dynamic loss of control, digital signal processing, the physics of undersea currents, the forensics of the recovered bodies, will, in the end of day, be "potentially distressing to "interested parties"" _ may them be the relatives of the pax, the relatives of the crew, the crew of L´Emmeraude, the CEO of Airbus and Air France, the brazilian coroners, etc. So I have no problem in define as "hypocritical" any tentative of censorship in this basis, and above all from the very person who once wrote:

Anyway, the point I am trying to make is that input from lots of different people with different areas of expertise gives the best chance of producing good answers in a situation like this. And that it seems a little unfortunate if some people assume that other people have nothing worthwhile to contribute.

All of this fits well with Mr. JD-EE remarks also, of course.

Regards,

CS

[GarageYears]

Pinger spec:

Operating Frequency: 37.5 kHz, + 1 kHz
Operating Depth: 0 to 20,000 feet (6,096 meters)
Pulse Length: ≥ 9 ms
Pulse Repetition Rate: ≥ 0.9 pulses per second
Battery: Lithium P/N C362-04270-2, standard
Lithium P/N C362-04270-1, optional
Operating Life: ≥ 30 days with standard lithium battery
≥ 90 days with optional lithium battery
Battery Storage Life in Beacon: 6 years
Acoustic Output: ≥ 160.5 dB re 1 μPa @ 1 meter
Acoustic Output After 30 Days: ≥ 157.0 dB re 1 μPa @ 1 meter
(with standard lithium battery)
Acoustic Output After 90 Days: ≥ 157.0 dB re 1 μPa @ 1 meter
(with optional lithium battery)
Activation: Fresh or salt water immersion
Beam Pattern: 80% sphere
Case Size: 1.30 in. (3.30 cm) diameter
3.92 in (9.95 cm) long
Case Material: 7075 T6 aluminum
Weight: 6.7 ounces (190 grams) maximum

There is no defined detection range... but I found this:

Deep-water Black Box Retrieval - November 2009, Volume 13, Number 09 - Archive - Hydro International

This states the typical detection range would be 2-3km, with 4-5km under optimal conditions...

But wouldn't you want to use one (or several) of these:

SEA 00C2 Salvage Assets: MINIROVs - Standard Vehicle and Open Frame Vehicle

Come to think of I thought they did have access to a similar device...?

Regards, GY

[EDLB]

JD-EE you are correct that the FFT window needs to be about 100Hz for the 10ms pulses. A bit frequency shift you get anyway by doppler effect, since the towed array needs to move. However you should gain a lot of signal to noise if you know that there is a 10ms pulse train to look for you can coherent digitaly decode. It is more than only the S/N gain with the 100Hz filter which is comparable to a Morse code filter. You know what morse code to look for and this gives you the additional S/N like in GPS. So the time stability of the ping interval is equally important. I think that the ping frequency should be best compatible with the sensitivity of typical towed arrays from hunting subs. They have all the DSP power on board you need for this job. Make the pinger compatible with its capabilities.

[GarageYears]

Confused?

I have been trying to follow the discussion on filters and what not, w.r.t. to detection of the pingers, but believe the techniques being discussed are way off what you would do to pull such a signal from the noise....

Let me explain:

I work A LOT with aircraft sounds, no need to explain more, but the MOST powerful signal extraction technique I use to isolate specific discrete noise sources from the composite is what I can "profile subtraction" - I suspect there is a "real name" for this but the general process is to sample a long period of baseline noise (for example the cockpit ambient noise in steady cruise, no changing parameters - speed constant, alt constant, etc). This is then 'saved' as the profile. Then I take my wanted sample and "subtract" the profile from the sample... the result is the residual (depending on the FFT depth and bunch of other 'precision' factors), and the quality can be remarkable.

Example: One humorous example came when doing the process described above, when out of the background noise, completely inaudible on the original sample, was a conversation taking place beyond the closed cockpit door, of a rather saucy nature between two FAs... I won't repeat what was said, but the signal must have been 30-35dB below the primary signal level, with the result being perfectly intelligible.

Having once worked on submarine simulators in the past (20 years ago now...) and having been involved in some aspects of the sonar audio, I believe this is a technique that is routinely used.

Anyway, if I were processing the acoustic signals, that is what I would do to pull the pingers out of the background signal.

- GY

[good spark]

all this talk about pingers, things towed behind submarines etc scares me, recent events have confirmed that if you have something lost in deepwater get people who know what they are doing involved, woods hole guys take a toffee out the tin, french navy go stand in the corner



gs

[bearfoil]

good spark

I tend to agree, however the BEA had intense pressure from the git to find and resolve this tragedy.

That said, the fact remains BEA is an agency of Government, and as such cannot escape the pressures and culture of Government (bureaucratic) ways.

Government is not responsive to Supply/Demand, and frequently needs to re-invent its mission on a case basis. Which suggests flexibility, but this is not generally the case. The environment in Government models is geared to influences that are frequently at odds with goals.

The ARMY, NAVY, BEA are giant orgs that have their own cadence. Financials and authority, or command grids, are archaic, command strict obedience, and degrade efficiency. Now this is my conclusion, but I have found it to generally be accurate. At the very least, Government financials are rigid, and disincentivise quick, highly technical, or rare challenges.

How about two toffees for WHole and one for BEA, Navy ??

bear

[EDLB]

GarageYears,

if you know how the signal looks like you searching for, you get much more signal to noise than with only digital reducing the ambient noise. However the current frequency tolerance and ping time interval tolerance are too sloopy to use this technique effective. My point is, that this can be improved very easily without adding much cost, weigt and volume to the pinger. Think about it as you know the exact conversation you where looking for in your example beforehand.

That technique is the only reason that GPS works with relativ small receive antennas and low transmitt power in the satellites.

[RR_NDB]

Another possible model trying to explain debris field shape II

Hi,

In my post #876 i presented a model trying to explain the debris field shape.

Since that, analyzing all info posted (on possible a/c "attitudes" after 02:10:10) i put a similar model, still imagining a/c "heading" ENE before crashing.

I took into account:

Engines psn
1 wing/MLG PSN (only considered the one at 5 o´clock)
Nose up at crash
Left bank at crash
Yaw to right at crash
255/075 signature of most debris
Engines traveling mostly vertical to sea bed
The distance between engines and psn of RH MLG as due splash/disturbance
My own Hypothesis of events since 02:10:10
Feeling (pattern recognition) of debris shape and orientation (IMO, clearly not random)

Model follows:

(milliseconds between points)

- A/c hits water, tail first at high vertical speed
- With less horizontal speed and going ENE
- A/c eng #1 hit surface, detaches and it´s "core" dives near vertically
- A/c eng #2 detaches, travels a little bit further than #1 and dives near vertically
- LH MLG separates from rest of wing and dives to seabed
- RH MLG separates from rest of wing and dives to seabed
- Wings and fuselage disintegrates and its pieces "showers" traveling WSW
to seabed. It´s parts submerging accordingly it´s characteristics (mass
and geometry) thus creating the debris shape.


For the image:




Waiting further info. to confirm PSN of eng. #1 and RH MLG that could reinforce or eventually kill this models (the first and this one).

And eventual info on existence of sea currents capable to create the observed "debris trail"

Mac

PS

With the objective of try to figure out the possible horizontal "trajectory" when a/c hit water.

[jcjeant]

Hi,

With the objective of try to figure out the possible horizontal "trajectory" when a/c hit water.

Other than an academic point of view .. what is the importance of the "trajectory" when a/c hit water or a/c heading for the investigation of the accident ?
Maybe few seconds before the a/c impacted with water the "trajectory" or "heading" was different...
Seem's to me as a "random thing"

[HazelNuts39]

Originally Posted by Machinbird

For a mental picture, imagine dumping a bucket of dirty water into a clear lake. The initial interface between the two would look something like an inverted mushroom.

Your intriguing model kept lingering in my mind. If your bucket was filled with pebblestones instead of dirty water, would they 'mushroom'?

I would draw a circle with radius 8.5 m around each engine position. Then, on a 95 percentile basis, any line connecting a point in one circle with one in the other could represent the orientation of the airplane's transverse axis at impact.

Regards,
HN39

[HeathrowAirport]

Tonights Flight Recorder Transportation?

There are two flights from Cayennes tonight. SOCA-LFPO

Either @ 21:00z as AFR3507 an A343 with an ETA in Orly at 05z
Or @ 22:30z as FWI571 an A333 with an ETA in Orly at 0630z

Does anybody know which flight they are going to be on?

[MurphyWasRight]

EDLB

GarageYears,

if you know how the signal looks like you searching for, you get much more signal to noise than with only digital reducing the ambient noise. However the current frequency tolerance and ping time interval tolerance are too sloopy to use this technique effective. My point is, that this can be improved very easily without adding much cost, weigt and volume to the pinger. Think about it as you know the exact conversation you where looking for in your example beforehand.

That technique is the only reason that GPS works with relativ small receive antennas and low transmitt power in the satellites.

One way to think of this is with a visual analogy:
Imagine of a large number of vertical lines on a sheet of paper, of random width colors and spacing (noise).

Then add precisely spaced 0.1mm wide red lines every 10mm.

Take a second sheet of black paper and cut 0.1mm slots every 10mm.

Now slowly move this sheet over the top of the first sheet and plot the
spectrum (color) of the reflected light.

As the slots move over the precisily spaced red lines there will be a pronounced shift of the reflected spectrum from all colors (white/noise)
to red.
Assume a detector with very narrow optical filter that filters out all but the frequency (red) of interest and one can easily find (if they exist) the
red lines.
Once they are found other information can be deduced from where the slotted mask was at the peak.

This is a (very) simplified description of part of the correlation techniques used by GPS to extract signals from noise.

Note that if the lines are not precisily spaced the slots will need to be wider and the "gain" will be reduced.

Also if the color of the lines is variable/iunknown (doppler shift) the "red" filter will need to be wider so signal to noise will again be reduced.

This is one reason that "cold start" of GPS systems can be much longer than with a warm start where known approcimate location, time and up to current satelite data are avaialable.

In a warm start the system is able to predict where (in time/frequency) each satelites should be.

EDLBs point is that if precision oscilators were used the frequency and spacing of the pings would be tightly controlled so the slot and filter could be much narrower allowing much higher gain.

(As I recall there is also additional gain if the frequency and ping rate are correlated)

A bit more precsion could be obtained if the frequency/temperature plot of the oscilator was well known and optimised for likely deep sea temperatures.
In case of deepwater one value easily determined is the likley temperature of the pinger.

BTW: Scanning too wide a range of slot spacings and colors can "find"
signals in random noise unless a very long sample is available, that could be what happened with the re-analysis of the sub data.

[tubby linton]

Two Paris judges rule that no more bodies will be recovered:
Air France crash victims 'must not be recovered' - Telegraph

[fantom]

Two Paris judges rule that no more bodies will be recovered:

Good news indeed.

[oldengmkr]

Takata,

In response to you comment: "There was no "reduced crew on the flight deck" either that we are aware of! Pilots were supposed to be at their post as per SOP." My earlier remark was made in the context of the much discussed issue in this thread regarding the recovery of the body of the Captain and "speculation" he was not on the flight deck at the time of the accident as well as the AAIB report remarks on the A340 as follows: "Consequently, the crew should have heard both warnings but may have assimilated both. ----- Therefore, the succession of aural warnings could have affected the crew's ability to assimilate concurrent synthetic voice warnings." If such an event unfolded for AF 447 perhaps if the Captain had been on the flight deck at the time the "entire crew" might have been better able to handle the circumstance. This is not intended as a criticism of any of the crews action. It is intended to simply point out the potential for the crew missing information during such an event that occurred on the A340 with which more individuals "on deck" might have been able to cope with if it happened on AF 447.

In regard to your other comment, there have been many posts in this thread since it was made which address the subject you raised and I have nothing to add to them.

HN39,
I also appreciated the AAIB re-releasing the report with the appendices. Comparing the A330 to the A340 plots, I find it quite interesting how two very similar aircraft in such close proximity could respond so differently. Perhaps this was due to different crews reacting differently - but it seems the AAIB doesn't think so based on it's statements. Perhaps it was '--- a random event driven by the severity of the turbulence." Or perhaps it is a control deficiency that emerges in a highly unusual combination of events that occur rarely - say every 9 years or so. If so, finding and resolving such a condition is essential, as the periodic loss of an aircraft (or two) is not acceptable to anyone - regardless of the cost of the effort. If the AF 447 data reveals similar behavior to the A340 then perhaps the AAIB should reopen the A340 investigation and conduct it in a co-operative effort with the BEA. Answers to both events must be found.

[gums]

Pulses and chips and...

Salute!

Murphy is getting to where I was intending.

The pulse length of approx 10 msec means that we have plenty of time chips ( the bits that Murphy describes) to encode the pulse. Same as the GPS encoding system that allows numerous satellites to transmit on the same freq. So we can transmit 3750 chips per pulse. If we only transmit once per second, then we can get by with very low power requirements on the part of the transmitter.

We know the base frequency and can scan back and forth until we get a distinct correlation if we know the "code". The signal can picked outta noise at really low S/N rations.

So why not specific codes for specific pingers?

Further, using only a few chips of the transmitted ping, we have a few thousand chips left over to transmit stuff like last known position and other stuff.

Gums sends....

PS, As with Murphy, I had to learn about GPS signal processing in order to develop our latest guided weapons and the aircraft interfaces and other stuff. For ground surveys we actually locked on to the base frequency of the satellites and with a phase resolver could get centimeter or even millimeter accuracy.

[mm43]

BEA and applied psychology

Remember when the BEA announced that the AF447 debris field had been located in the Abyssal Plain and its exact position would not be released?

Well accompanying that press release was a graphic with the a very large arrow pointing to a position near the eastern side of the Abyssal Plain, and the end point on that arrow actually marked the true position of the debris field.

Machaca made a composite graphic at the time using a BEA graphic, previously modified by me, overlaid with the latest one from the BEA.

It would appear that a much practiced form of applied psychology was inherent in that press release. One would have to wonder who was teaching who - Osama Bin Laden or the BEA?? If that was not meant to be the case, then apologies to the graphic artist who has without doubt by now lost his/her job.