Recorder Locations (distributed memories)
 

There is obviously some merit to rethinking the CVR/FDR data recorder concept. One proof point is (if I understand correctly) that BEA is indeed searching also for the other memory units in the aircraft, i.e. within the flight deck and engine FADEC systems, even though they are not hardened to the extent of the CVR/FDR memory units.

However, a simplistic relocation / multiplication of units brings its own problems.

- The added dead weight and space requirement of each unit and its power feed. As noted before, it performs a critical function at most only a few times in the lifespan of the aircraft.

- Each unit needs a reliable incoming data feed with all recorded parameters. For obvious reasons it would be impractical to feed each recorder with primary data (i.e. individual electrical/pitot/etc. sensor lines). This implies a bus structure and furthermore a need for multiple data buses, otherwise the "single point of failure" of the recording system just shifts from the CVR/FDR to the bus multiplexer equipment. Multiple buses and multiplexers add more weight.

- Increased amount of wiring. The data buses could probably be light optical fibers, but even so I presume there is usually not much extra wiring space in an aircraft. So the installation probably needs to be considered at the basic design phase. This means the time to deployment is long.

- An option might be to have wireless transmission of the data to the recorders, but that brings with it some radio spectrum management issues. To exaggerate a little, you probably don't want the recorders to be susceptible to your seat neighbour's computer Wi-Fi. (I remember Gulfstream has done some trials of using wireless links - IIRC for spoiler actuators - but this is not a trivial issue).

- Looking at recent crashes and the CVR/FDR searches, in my mind the eventual success rate of finding the CVR/FDR in a usable state and extracting the data has been remarkably high. The main problem has been to find the actual aircraft wreck - once it is found the recorders have also eventually been found fairly quickly, actually in each recent case that comes to my mind.

- From the above it follows that a wiser way of spending R&D and development money, with a higher benefit, could be to improve pinger and CVR/FDR locating techniques, rather than to just add recording equipment to the airplanes.

Just my 2 cents.

Airbus A330-203, manufacturer serial number 660, registered as F-GZCP.
First flew on 25 February 2005.

The U.S. NTSB and TSB of Canada issued safety recommendations on March 9, 1999, to require:

By January 1, 2003, all newly manufactured airplanes required to carry both a CVR and FDR be fitted with two combined voice and data recorders, one recorder located as close to the cockpit as practical and the other as far aft as practical.

@ DJ77 post #582:
Thank you Sir, much appreciated :)
Seems a good idea to drop & search for another ULB, while they're at it as you said.

@ Bizman posts #572 & #621:
1/ Seen from here, english is (still) the first link on each item since the Apr 29th briefing on the english page. French is first link on every item on the french page.
2/ What's the point ?

@ DozyWannabe post #624 :
Just quibbling, perhaps, but the Habsheim crash killed 3.
/end of off-topic and :ok: for your post.

@ MJC2 post #628 :
+1 !
& thanks for the digging work :)
I assume the original mis-interpretation (also) came for the bad translation of interim report #1 form french to english (at last on this sentence, but there are more examples of "clearly french to english bias" in the report).

@ all, about the recorders reseach, location on the aicraft, technology and so on : you may find it's worth (if not already done) to read the Report from the International Working Group on Flight Data Recovery, 22 December 2009 published by the BEA on its website.
Direct links : French version / English version.

No power, no data
 

Caygill,

The solution must be based in:

"Fault tolerance and graceful degradation".

In this case the idea would be allow a "minimum of information" to allow investigators understand the facts until the sea surface impact.

How it would be designed? We need to prepare an Specification on "what we need".

Obviously in an extreme "electrical power loss" in the a/c systems you may ask: Why record?

Unless you distribute also some sensors.

There is an option for the SSFDR, the accelerometer. That could work integrated to distributed sensors, and that could be made "power independent" of aircraft power during some few minutes.

This is a "System Issue" and IMO there is room for improvement.

What in my opinion is not acceptable is:

Spend years to start (to try) understand what led to this loss.

RRNDB, do your design aircraft? I hope not or I am sure that commercial aircraft accidents will happen weekly. Maybe you should start a new thread for flight recorders. Graybeard has answered your recommendations adequately, so let's move back to this thread, and the reason for it.

Something should be done
 

snowfalcon2,

I agree. A good start point!

How many hull-loss accidents have NOT been solved as a result of unusable/unrecoverable flight recorders? It can't be more than 5. So hand-wringing about sensors is pretty much pointless if the actual goal is flight safety. The introduction of CVRs and FDRs has increased airline safety by an order of magnitude - only another order of magnitude improvement would justify changing the whole system. I think good training with emphasis on flying skills is much more important than obsessing over recorders and sensors. I get the strong feeling that the bad man of AF447 is the flight management system and the over-reliance on automation to fly the airplane.

RR NDB – ref your Tail Fin separation (before or at sea surface crash?) post ... #593
 

RR NDB:

There has been much interesting discussion on these pages about whether a large airliner can be kept flying in a controlled manner minus a VS.

Shown below is a 1961 picture of an RAF 74(F) Sqn English Electric Lightning Mk1. It has a 60 degree swept wing and a low set slab tailplane.

It landed uneventfully with a considerable amount of VS (vertical stab / fin) missing.
The structural failure was caused by inter-acting shockwaves from adjacent aircraft.

The pilot, Flt Lt Jim Burns, was the outside man in a 'finger four' low level air display high speed formation pass. As the formation of 4 aircraft crossed RAF Coltishall airfield at Mach 0.95/620 knots IAS and 200' altitude the outside aircraft pulled up abruptly and then made a leisurely circuit. It landed minus about 75% of it's vertical tail in an otherwise normal configuration of full flap and a Vref of 165 knots. The pilot was unaware of the VS loss until he got out of his cockpit: his VHF radio antenna was at the top of the fin and had also departed –

– so no radio, and yes it was VHF when this happened 5o years ago.

http://i31.photobucket.com/albums/c3...rnsfin3453.jpg

I think that is the point RR_NDB wants to bring forward :
If the recorders have their own independent power supply, they will keep recording.

This has already made the object of a recommendation following the Azores glider as the last 20 minutes of data are inexistent :

rubberband2

That is a great pic.... It supports directional stability w/o Rudder, but it appears that the Fin is mostly present. The Rudder and the Vertical Stabilizer are responsible for two different aspects of flight.

The Rudder is basically a trimming device, the VS provides directional stability. It can be said, therefore, that the Rudder's task is the opposite of the VS, it stabilizes Yaw, in the turn, or establishes a crab for offset.

An a/c can fly quite well without the Rudder, as your pic shows. An a/c is not airworthy w/o the Fin.

CONF iture;
If that is RR_NDB's point, then he hasn't thought it through very well. He hasn't said what parameters the recorders will keep on recording and how that works.

The only suggestion of independent power supply that has emerged was after SR111 and that was the voice recorder.

PJ2

Loads?? Direction of Flight?? GPS (altitude and trajectory) ?? Wouldn't that be most important in the post mortem, the sequence of loads on a passive a/c??

Obviously, current boxes wouldn't record a thing, but you could add some gyros to them to get them record accelerations... Some aircraft used to have a separate simple pitot-static system in the bum for their rudder travel limiters... could easily be done to feed an FDR. ...But why?

If the ship is completely dead, electrically, what a self-powered FDR might record (on a FBW aircraft!) beyond that point is totally predictable anyway, isn't it?

(BTW... "West of the Moon" I get, since the Moon always shows us the same side... But "East of the sun"? Yer 'aving us on! ;))

Hi,

Concretions on the CVR pinger ..... indication of a crack ?

http://www.crashdehabsheim.net/autre...DFDR/cvr-2.jpg

http://www.crashdehabsheim.net/autre...DFDR/cvr-3.jpg

Were Load traces supplied in the report Perpignan??

I see what looks to be corrosion oozing from those black keeper brackets, but the image isn't quite clear enough to be sure of much...

Brief note re the discussion about a/c continuing to fly with part of the vertical tail missing....

As an aeronautical engineer, I wondered a few times, why todays airliners with two wing-mounted engines have such huge vertical tails, compared to other aircraft.
The answer turned out to be, that the 'design case' is directional stability and control during an engine failure at low speed (i.e., during take-off or landing).
In most other 'design cases' a much smaller vertical tail would be adequate, as demonstrated by that B-52...

Even without a vertical tail, the rear end of the fuselage provides some "weathervane" stability, but in that case the complete lack of a rudder can be a nuisance.....

bear, the report is on the BEA site.
It's some time since I last read it, so, I'm sorry, you'll have to do your own research.

Of less importance than OEO performance is the fact that some twins are "short coupled", meaning the tail's "Arm" has less effect than an a/c with a longer fuse twixt flap and tail cone. The 777 looks much "better" (balanced) than the guppy, for this reason, imo.

bear

My point of logic was: if you have no power for instruments and sensors there is absolutely nothing to record. :ugh:

Sure, you can verify that any specific data was missing, but if the data cannot be generated or transferred, it cannot be recorded. If you loose all power on a modern aircraft there is nothing left to record if you do not create a completely independent array of sensors powered equally independently

@jcjeant
I think this may be galvanic action where the aluminium housing of the pinger is acting as an anode and is corroding rather than the ferrous cylinder. Further evidence of this is that there appears far less rust on the CVR cylinder than the FDR which has lost its pinger, and hence no sacrificial anode to stop it rusting. There needs to be an electrical path between pinger and cylinder for the effect to occur, which I assume is there somewhere.

Galvanic anode - Wikipedia, the free encyclopedia

Sacrificial protection on the hull of a ship using a Zinc anode.

Zinc anodes protect the steel parts of the hull | Flickr - Photo Sharing!

bearfoil,
What's a 'guppy' in your neck of the woods?

In my world, the 'Guppies' are blown-up Stratocruisers and A300s with faulty cabin pressure regulators.... LOL.

Airbus CVR designs currently record 4-channels - the Capt, F/O and 3rd occupant crew mic signals, and the CVR area mic. Since the crew microphones within modern aircraft are all 'powered' (electret), these will be lost if electrical power is lost. The dedicated CVR microphone could be powered from an internal supply sourced from a battery supply, so that would remain active.

Might be an American or 'company' thing, Chris, a guppy is a 737. Alas, they are all gone, victims of the 320. (Not that there's anything wrong with that!!)

:ok:

Corrosion on pingers ?
 

If it's corrosion on the pingers, it is probably galvanic corrosion ( occurs between metals of different potentials in seawater).
But since the waranty ends after 30 days at sea, I don't think the designers cared about this.

Modern Airliners are not realy designed like a "weathervane". Large part of the fusalage is ahead of the CG as opposed to e.g. the B52. The lateral drag of the fuselage that incurs ahead of CG generates a destabilizing yaw moment, the lateral drag aft of CG generates a stabilizing moment. I.e. the larger the part of the fusalage aft of CG, the better the a/c will fly without VS.

A swept wing without a fuselage doesn't need a VS to fly:
Horten Nurflugels

Caygill,
"If you lose all power on a modern aircraft..."
Rare... since there are batteries, and inverters to generate some essential AC from the 28V DC battery bus.

Us engineers are not quite as stupid as you seem to think.
RTFD... and so far I've seen almost nothing about the A330 system.

RR_NDB has a few points.

OleOle


Define "Fly"......

And what tells you that's not already the case?
Unless you have the full WDM for the A330 at hand....

Something like seen here ?

YouTube - Horten Ho-2 Flying Wing Test Flight 1935

[offtopic]
OK, bear, never heard that term for the 737...
Saw 'Fat Albert' a few times, then found that that particular 'sobriquet' had shifted to the C-130....
I like the 737, less of a noisy flying Coke can than the A320 family, from the SLF point of view.
[/offtopic]

OleOle

I flew a Manta 'Wing' in the seventies, so am familiar. Just a little fun poked at Jack Northrop and his obsession. At a billion dollars a copy, the B-2 is less a flying wing than an airborne Bank, or money sewer, depending on one's pov.

:ok:

ChristiaanJ

Then there is the BUFF. Boeing gets the familiars, eh??

Which I do...

18V DC is derived for the CVR area mic preamp, from the 115V AC supply delivered to the CVR.

Trying to keep some VS effect for a minimum of "airworthiness "
 

ChristiaanJ,

The separation of VS in both cases (AA587 and AF447) occurred at it´s "coupling" to the fuselage. This "coupling" in EADS design appears to be the "point" that fail when the loads exceed the specs. The "connection" between "conventional metal" to "new materials" advanced VS structure.

Question:

Wouldn´t be better/safer to have a "tapered design" with greater "Fault Tolerance" and "Graceful Degradation".

I mean:

Considering the main reason for a big VS is for take off * why not to have a tapered (strength) design?

To keep some VS effect even after exceeding specs limits (due any reason)

(*) Exceptions: Windy and cross wind Landing

Rationale:

Redundancy, fault tolerance and graceful degradation are "defense lines" against our "nightmare": Murphy´s law.

http://www.ntsb.gov/events/2001/aa587/AA587_09.jpg

http://www.ntsb.gov/events/2001/aa587/AA587_10.jpg

Mac, you made your point.
Once the entire VS comes off, all bets are off too, one could say.

Very long New York Times article on the finding of AF 447.

What Happened to Air France Flight 447? - NYTimes.com

There is a lot of history, and some theorizing. Quite a bit of information on the autopsies, with the doctor indicating that he couldn't rule out that some might have survived the initial impact.

RR NDB

"Sequential Failure". A time worn concept, whether "crumple zone", or ablative skin.

The problem is at least "twofold". One is the almost unavoidable problem of mating plastic to metal, mechanically (1), or bonded (2).

On the Airbus VS it is mechanical. This presents the "focal energy" obstacle. A concentration of stress in Epoxy/Reinforced structures is counter intuitive to the material's strong point. Thus a dowel, pin, or clevis, bracket system focuses failure sums of energy on an assembly whose design is related to stress/spread for maximum load areas. Note the A350 has gone to 5 (five) saddle/pin systems.

The second, bonded, is universally avoided unless some spread can be attained.

The most important consideration is the fact that this most critical assembly is "surface mounted". This concentrates virtually all Stresses on an inescapably small area. The VS/Rudder is vertically cantilevered, a swell looking and very efficient way to alleviate drag from external wetted areas.

However.......


edit... I cannot locate the "lateral rods" on the bare aft fuse of the 587 A300. These were added to later a/c....... There are pictures of these rods on the VS (447). After the fact design, hmmm.......

edit... Disregard, I did find a remnant of the "Lateral Rod" on aft bracket.
sorry...

No it didn't. In AA587 as shown in your images, the VS structure was torn off the couplings.

In contrast, in AF447 the couplings did not fail and were still attached to the VS along with bits of the fuselage. The rear fuselage itself failed. BEA first report has images (P39): http://www.bea.aero/docspa/2009/f-cp...90601e1.en.pdf

Bear

Think you got your numbers mixed up, it isn't a 737, it is a 377 Super Guppy:ok:

http://upload.wikimedia.org/wikipedi...SA_landing.jpg

Precisely! One shows a fuselage with bits of fin attached, and the other shows a fin with bits of fuselage attached. Huge difference!