Think I need to bite at this one, although my better conscience tells me not to.

1) We are long past the days of "tape" in a modern flight recorder.

2) In the whole history of aviation there have only been a handful of broken or irrecoverable recorders.

3) To transmit data every 15 minutes is hopeless, It is usually the last 15 minutes of data that are the most important and under this ludicrous idea that would all be lost.

4) Have you any idea of how much data transfer via satellite link really costs? Most of this huge amount of data would be totally useless.

5) Do you have any idea how it would be possible to maintain an uplink in severe weather conditions and unusual aircraft angles?

6) Yes we do use the internet to transfer data, predominantly via cables/fibre networks and after that via GSM networks, I don't recall passing any GSM towers in the mid Atlantic, but possibly I should look more carefully next time.

7) Please read all the other post regarding this point.

tex

Thanks Tex37!
Makes my answer simpler!

:D :D
It just so happens that, before retiring, one of the last things I designed and got to work was an ATE for the FDRs we manufactured.

Your electronic and IT experience clearly ended with the Commodore 64 and programs on cassette tape.

CJ

Has there ever been a cost/benefit study of CVR and FDR? Hauling 50-100 lb of hardware and wire for the lifetime of every airplane is no small cost.

GB

Interesting question, GB ....

What is the cost? Fuel to haul the equivalent of a couple of extra suitcases and some duty-free, or one more empty seat, for the lifetime of the aircraft, plus the cost of maintaining the CVR and FDR.
Less than the cost of carrying diversionary fuel, which is mostly not used... to put the weight in perspective.

What is the benefit? More difficult, because it is very difficult to assess how much and how often CVR/FDR data from a given crash have prevented further crashes (hence more hull losses and more loss of life).

Almost worth a separate thread, to avoid "polluting" this one.

CJ

ChristiaanJ
It should be quite possible to scope the total bandwidth that would be required for a continuous uplink system to cover e.g. the Atlantic. I'm sure that it has already been done many times and that the answer is not currently feasible, but I haven't seen it documented.

From your experience with FDR's, what is the peak data rate that a typical large aircraft records?

The French are "confident" that the recorders will be found, this new article says, but the reporter "could" be slightly wrong about the search area:

France confident of finding Air France black boxes - thestar.com

Square kilometers would be better :}

I was thinking about this yesterday. For crashes on land where the a/c remains relatively intact (or in one place) the current FDR designs work pretty well - they're hardened and are relatively easy to locate.

What is needed is data redundancy for when the ADR can't be found. An uplink would be a not very reliable, not very high bandwidth, not very cheap form of it.

A better idea would be to have 2 or 3 small auxiliary hardened FDR memory units which are slaved off the main FDR and attached to pieces of the aircraft that are more likely to float or be readily recovered in the event of a water impact/breakup. e.g. the tail assembly, or even attached to a float. They wouldn't have to be as large (they don't need any of the input processing etc.) or as hardened as the main FDR, just watertight and impact resistant enough to survive e.g. a water impact.

This is not a comment on your idea as such.
But most present-day FDRs don't do any input processing.
Their input processing is done by a separate FDAU (flight data acquisition unit) which then sends the data in digital format (ARINC 717, usually) to the hardened FDR.

CJ

My experience dates a few years, so I did a bit of "informed Googling"...
The data rate tends to vary widely, depending on the age and the size of the aircraft.
64 and 128 12 bit words per second sounds familiar to me, but it's now up to 256 wps, with 1024 wps on the horizon.
That translates to about 12 kbps raw data, and for robust transmission via satellite I would say you would have to double that (more checksums, more labels, more sync).

Doesn't sound too bad, until you multiply it by the number of aircraft in the air at any one time (about 10000, of which roughly 5000 over the US, so presumably several hundred over the Atlantic).

Not to mention there is no infrastructure to handle any of this sort of data, or assure a continuous and robust second-by-second data stream...
As mentioned already, packets every 15 seconds are useless, a lot happens in 15 seconds.

Also there is no mention of the CVR in this context, which would easily account for another 12 kbps for the recording to be of any use.

CJ

Thank you Christiaan.
So to build a system that is reasonably future-proof you will need 100kbps per aircraft, 1000 aircraft over the Atlantic, total 100 Megabits.

The challenge will be to design and build a robust satellite-based TDMA system that can control and synchronise the individual transmissions. And an entirely new radio fit in all aircraft (with appropriate redundancy of course because the radio link will always be less reliable than an on-board recorder).

It's not impossible - given vast resources and general agreement from all the relevant national and international organisations - none of which is about to happen any time soon.

I think this puts the idea to bed.

Of course 100Mbs is nothing... it's about one 2nd-generation Ethernet link, no?
So when we talked about just bandwidth, we were barking up the wrong tree.
"The truth is elsewhere"...

You already ponted to some of the other issues...
And a huge infrastructure on the ground everywhere, to recover these data, save them, distribute them, etc. etc.

I just used my imagination a moment...
Can little "AZ Airlines", trying to operate half a dozen second-hand regional jets conscientiously on a shoe-string budget (yes, they exist and try to survive), even afford to subscribe to such a service (they never could do it on their own)? Or can they be relied on to have the facilities to receive and store the data reliably?
Their half-dozen aircraft will carry FDRs, they have to, and they're likely to maintain them as they do the rest of their equipment.
Not to mention that if they're clever, they have already tied them in with QARs etc.
What is their interest in a system that "maybe" recovers a few more data from an extremely rare accident, where the FDR itself is not recovered?????

I can think of many more reasons..... like the one already mentioned... the posited "data link" breaking down because the aircraft is in a vertical dive, for instance.

The idea mentioned by XB70_Valkyrie would be nice... were it not that a few more FDR memory units would be equally impossible to locate as the original wreckage.

Personally, I still like the dye markers.... and the improvements suggested to the ELBs and ULBs, and possibly adding more of those.

Contray to the suggestions about adding FDR units (which would mean major aircraft system modifications), they would mean minor, and largely system-independent and aircraft-independent, improvements.

CJ

Unfortunately, no.
100Mbs from one aircraft would be easy.
But we don't have that, we have 1,000 separate data streams sent up to the satellite receiver. The transmissions can't be sent up blind from each aircraft, they would have to be synchronised and controlled from the satellite base station just as mobile phones are controlled by their network. It's a much more complex requirement. And synchronising high-speed data over long and variable-length paths from fast-moving vehicles is very difficult. Not quite impossible, but extremely difficult to do with the required 99.9999% reliability.
Only then can you combine all these data received in the satellite into a single stream to send down to the ground stations for distribution to the surface infrastructure you correctly identified.

edit: I still like the dye markers too!

Sallyann1234
I may not have been 100% clear in my reply.
But I think we're already 99.9999% on the same wavelength.

CJ

Perhaps even quasi-synchronous :)

A satellite uplink is only needed when there is just one aircraft in the neighbourhood. In this case the aggregate data rate for the neighbourhood is that of a single aircraft. Where there are multiple aircraft then each can broadcast its data to its neighbours. Since it is unlikely that the neighbouring aircraft would also crash, they do not need to keep the received data in hardened memory; they can store it on a regular PC and upload it to the Internet on arrival.

Arguably, the only really vital data to be transmitted are the GPS coordinates. If these were updated every 5 seconds then locating the crash site would be easy and the rest of the data could be retrieved from the black boxes. If the data were available in real time survivors could be found quicker too. The badwidth required would be much lower.

However, as already noted higher performance pingers would be a much simpler solution.

As already said, the bandwidth is not the key problem.
Your solution too would need a brand-new, very elaborate and very robust infrastructure, which currently doesn't exist.
As indeed already mentioned in the latest DGAC report.

CJ

The above started me daydreaming of a simple system.

Could one have a "buddy system", where each aircraft is paired with a buddy? i.e. two aircraft on the same or nearby routes maybe 10 to 30 minutes away from each other are "paired". Each aircraft has its own DFDR as normal, but each also transmits its full DFDR data stream to its buddy, which records it.

On landing and confirming that its buddy is safely on the ground, the received data is deleted. Gone.

Advantages include:

(a) No other infrastructure is required. No satellites or central data handling is needed. No massive amounts of data is stored. The cost is distributed between all airlines and countries

(b) The full data stream of the downed aircraft is available immediately. Also, it might be possible to transmit just the last known position to emergency services even before the receiving aircraft lands.

(c) While I hesitate to mention this, it does cross my mind that the buddy aircraft will most times be from a different airline and maybe a different country, minimizing the possibility of any funny business in withholding, delaying, massaging, or losing the data.

There will some routes where there will be no buddy within range, but maybe only 1% of routes.

Just daydreaming. :) http://pickyperkins.home.infionline.net/pi.gif

PickyPerkins,

Think it through.....

- There is no equipment in current aircraft capable of transmitting, or receiving and recording, a continuous 10 to 20 kb/sec data stream over a distance of a few hundred miles, omni-directionally.

- The FDR data in the aircraft (usually between FDAU and FDR) are perfectly useless for broadcasting... they would need additional time-stamps, sync, labels, etc. .... i.e., your scheme would require another piece of equipment for the re-coding, which currently doesn't even exist.

I can think of several other points why your scheme is impractical....

Only incremental improvements of existing equipment (such as the pingers) or very cheap add-ons (such as the dye markers) have any chance of being accepted to solve what is a very rare problem.

CJ

4000m is a long way down. In terms of design modifications to achieve retention and transmission of information, it's all fixable ... for a price. Wreckage location: simple buoy to be released following g measurement exceeding preset level (crash). Buoy contains GPS receiver/data logger and beacon, retrieve buoy and download data logger to locate release point. ELF receiver to trigger further buoy release (assuming USN ELF system is still in operation?), ELF signal to be sector specific (but I doubt if ELF can reach to 4000m). Buoys may contain flight data loggers, data to be stored on (several) SD cards or similar.

All adds weight/cost.

Mike,
Nearly all your suggestions mean new and currently non-existing equipment.

Yes, this has been suggested and wouldn't cost that much. Sticking a few more, and more sophisticated, ELBs (emergency location beacons) on the airframe.

But it's still as I said: "new and currently non-existing equipment", and the next aircraft going down will just happen to not yet have the equipment installed....

And no, ELF doesn't work that far down.

CJ

This thread is heading for banishment to Tech Log, so I might as well contribute to its demise.

The kind of data the FDR records is called State of Health data, on spacecraft at least. It is generally boringly repetitious and compresses nicely (losslessly, of course) sometimes using algorithms that insert values (for each channel, or parameter) only when they change by a meaningful amount. More dynamic parameters may be recorded by inserting differences between consecutive values into the data stream, each difference taking fewer bits than stating the whole value would. There's a popular silicon-based encoder (popularly called Rice encoding after one of the designers) that choses the best algorithm on the fly, no pun intended. And all this stuff exists in boxes with 1553 interfaces, since that's become popular at higher altitudes, so to speak.

I don't look at FDR data so am guessing, but would be pretty surprised if 10:1 compression wasn't a piece of cake. The costly bit would be making the box and uplink robust enough to survive a nearby explosion long enough to deliver the last message. While the normal data rate would be low, a serious event would generate full values for each parameter at the system's normal sample rate, for the duration of the event, and that would have to be preserved and uplinked perhaps over a period of several seconds.

CVR data needn't be lossless. I have nothing special to contribute in that arena, but we all realize that a minute of good-fidelity music becomes something like 1MB of mp3 data. Voice quality would be much less than that. As with FDR data, in normal flight there would be next to nothing to send.

I know nothing about what happens next, but note that Hughes and others operate geosync spacecraft with transponders dedicated to providing internet service to hopefully more than 500 customers (cited earlier as the number of transoceanic flights in some area). They're happy to support megabyte data rates for their users - far greater than needed except for mayday situations. Afaik no such service exists in the right place for oceanic monitoring, but that doesn't mean one of the maritime birds wouldn't mind providing such a capability. Hughes gets around USD80 per month per user, btw, and per the TOS doesn't appear to get annoyed until users start using hundreds of MB in a short period. I'm assuming that uplink and downlink are symmetrical for them, unless they do a lot of common-data caching on board the spacecraft. That's unlikely.

What to do with the data seems to bother people. Clearly the system would be operated by a central authority like ARINC, who could simply store it in a disk array at the satellite ground station - a giant FDR.

The thing I haven't mentioned is the uplink, because I truly know zilch about that. Hughes/Directway uplink antennae tend to be dishes around a meter dia, I think.

Has it been confirmed that one FO, recent to the type, was alone in the cockpit at the time of the incident?

That can't possibly be confirmed, ever, unless the CVR is found and there happens to be a message from whoever was on the FD speaking aloud "Well, it's just me on my ownsome" or similar......

Payscale,
Seems highly unlikely, given that they were heading into some nasty weather.
Even the least-experienced flying crew member had 807 hrs on type and nearly 3000 hrs TT.
Rockhound

The true engineering spirit
 

Which in itself is absolutely no excuse not to start working on a proper solution. I find it hard to accept that with the technology available today it seems nearly impossible to locate an airliner that went down into the ocean.

I also do not like the argument that a solution is not needed because "it does not happen very often". This accident could be a landmark accident in relation to the use of pitot tubes in high altitude adverse weather. Also: think about what happened to the BA 777 at Heathrow. Does the fact that the apparant ice accumulation in the oil-fuel heat exchanger only brought down one airliner make it any less important?

No-one else likes it either.

But in a world where airlines are struggling to survive, it is simply unrealistic to expect them to spend huge amounts of capital and further commit to ongoing loss of revenue, in order to recover data from a one-in-a-billion accident.
It is even less realistic to expect agreement from airlines and administrations across the world on a single course of action. Their priorities are elsewhere at the moment, as can be seen in the headlines.

Not sure if its been covered but given the extreme depths that are likely involved the dye suggested may never actually make to the surface let alone anywhere near the prang.
There are thermal layers that affect acoustics I believe, so a dye solution may never appear .OK at shallows probably but them you would likely find the recorders anyway.It would certainly be helpful in many cases.
Anyone seen smoke from a large fire on a very calm day , sometimes the smoke appears to hit a glass ceiling and commence to spread out?

Given many airliners now use solid state recorders ie the fairchild 2100- series that are much lighter than the old models how about just adding to battery capacity ? All solutions will cost but suspect adding capacity in future maybe a relatively cheap option.

Live info squirts via satellite / acars for example maybe ok if they are only triggered by an event , ie engine out , manual ops by crew, excessive rate of descent, excessive g forces, loss of cab pressure etc.
Otherwise just too much info & too many flights to record.Much less to record / monitor if only triggered by an event.

Just brainstorming..

There´s no reason to abandon the present recorders for a wireless system since onboard recorders already are mandatory and the cost of having them already is calculated into the business model. Just have to make them easier to find, just like is being said here.
The Boeing 787 will have enhanced recorders (CVR plus FDR in a combined unit, times two) which will record many times the data of previous recorder models and one of them will have a ten minute independent power supply. The "only" problem is, they are designed differently and collect data directly from sensors in a fiber optics ethernet data stream, which makes it impossible to retrofit these enhanced recorders to older aircraft types. Read about it here:

http://dev.flightsafety.org/asw/jan0...n08_p47-48.pdf

Hi Finn,
I was already becoming aware of ARINC 767 (mostly through this thread, I readily admit!).
However, ARINC 767 doesn't yet seem to address finding "lost" FDRs...
Or does it? The article you linked doesn't mention it.

Having two FDRs at 4000m under the ocean, rather than one, doesn't improve matters, if you can't find either of them....

CJ

Having two units, each with more data than ever before obviously will be a great improvement as such, but the French investigators seem to be the first to consider finding the boxes a problem worth looking into. Maybe there should be two improved ULB beacons per recorder? Provided there is enough physical free space available where the recorders are installed in the aircraft, of course.

Can I just put the satellite link to bed once and for all.

There will be noway of transmitting data up to a satellite and down to an SGS (Satellite Ground Station) for recovery or use in an inquiry. Yes, the technology exists, but there are just not enough satellites up in the sky to make any sort of inroads into the number of channels that would be required. I suspect nobody will put satellites up on behalf of the air industry, so they would have to fund it themselves. On average it costs about $500m to put a satellite up into orbit and maybes 200 more may be required up there. That cost (expensive even for airline budgets) would have to be passed onto the passengers, which would make flying extremely expensive.

Secondly, placing an omni-directional (which it would have to be) satellite dish on board and aircraft would cause an immense amount of pain for any aircraft manufacturer. Also, you cant hop from satellite to satellite (like mobile phone coverage) you need to know the specific upload/download speeds and upload/download freqs for any data transmission. Any satellite on board an aircraft would have to have auto-tracking and know exactly where and when it should be looking in the sky for a satellite. Most satellites up in the sky move around and do not stay in the same location, so the aircraft would have to constantly be looking for that movement, as well as counter acting its own movement.

Yes, info could be transmitted blind, hoping the satellite would pick it up, but what would be the point in that!

Rockhound, according to the latest interim report, the actual whereabouts of the Captain cannot be determined. His body was recovered and autopsied.

Perhaps a bit more strange, none of the several CC seats recovered were occupied at the time of impact.

I thought we already did?

Bandwidth as such isn't even the real problem, as we calculated earlier, it's the several thousands of individual little data streams, that have to be handled securely without any interruption. While technologically possible, the whole system would have to be designed and built from scratch.

As you say, the aerial is the other problem....
Satellite aerials are not omni-directional... even relatively wide-angle ones still need steering and locking on to the satellite.
Going down in an unusual attitude would most often mean losing the lock-on.
So would flying down a valley before crashing into a mountainside, unless the satellite happened to be right overhead.

So I agree with you.
Let's concentrate on finding the recorders themselves, come hell or high water (in the most literal sense...).

CJ

Satellite aerials are not omni-directional
Stricto senso this is incorrect.
Eg. Iridium, Orbcomm, Globalstar, Thuraya are satellites or constellations of satellites that are accessible by portable equipment with rubberduck type of antennas.
What matters is frequency band (L, C, Ku or Ka), what it is aiming for (in terms of bandwidth/speed, which correlates to satellite power and antennas' gains) and type of orbit and height (LEO, MEO or GEO, which correlates also to latency, a critical factor for some applications).

iakobos,
Stricto senso you are right.
Which is why I said it's "technologically possible".... I already was thinking of that.

But none of those systems are capable of handling thousands of 100 kbit/sec data streams simultaneously, reliably.

And try to hold the rubberduck (I am familiar with the type of antenna) upside down under an aircraft and see what ERP you get towards the satellite(s) you are relying on for those last critical seconds/minutes of data.


Something passed through my mind...

Twenty-five years ago, being able to flip open a little 'StarTrek'-like communicator open nearly anywhere in the world, an talk to nearly anybody else on the planet, would have seemed a pipe dream. (Admittedly, there are still places where you need a sat-phone, but I think you get my point.)
Now it's almost normal.

Twenty-five years from now, we'll probably expect to be able to plug in our little communicator on a plane anywhere, and instantly get connected to anybody else on the planet.
FDR, CVR and a lot of other data could ride piggy-back onto such a service.

But while FDRs and CVRs now are slowly and steadily becoming mandatory on ever smaller aircraft, even if not not on your Cessna, how many of those "data transfer" systems would in in place (and not part of the MEL) even then?

Let's stick with FDRs and CVRs (and VCRs if they can be shown to be useful) powered by the essential bus, that will stay with the disaster until the last possible second.

CJ

unsafe condition
 

The inconsistency of measured speeds resulting from blockage of Pitot probe is "an event that can cause victims usually with the destruction of the plane! This event can lead pilots to excessive workload which does not allow them to perform their tasks accurately or completely! “

The FAA (USA) made that affirmation several times in a document* dated September 9, 2009 using just 2 words : "unsafe condition"!

The BEA has very kindly referred to the definition of an "unsafe condition" in his report of last December 17.

What the FAA said, the European Agency for Aviation Safety (EASA) has not wanted or dared to say ... but saying it anyway.

When, in August 2009, the elimination of the Pitot probe Thales AA, fitted to the A-330 flight AF 447, is considered mandatory, EASA issued an "airworthiness directive” (AD) but claiming that it was simply a precautionary measure.

An AD for a precautionary measure, it does not exist!

When the manufacturer and EASA detect a problem that is not an "unsafe condition" but that requires a response, EASA publishes a SAFETY INFORMATION BULLETIN (SIB).

Extract of EASA document** entitled "Continuing Airworthiness of Type Design (CAP)" (March 2008). Page 39:

Only when design related issues which may lead to unsafe condition are considered likely to exist or develop, issuance of an airworthiness directive is warranted.

And

Information may be available to EASA related to airworthiness concerns on aircraft under national registers, but for which insufficient evidence exists to qualify this as an 'unsafe condition'. In such a case, the PCM [1] may elect the publication of an SIB, containing information for the safe operation of the affected aircraft.

To remove the Pitot probe Thales AA, EASA issued an AD… not a SIB.

There was therefore an "unsafe condition" that required an answer (the elimination of the Pitot probe Thales AA) !

This response came very late because the pilots of flight AF 447 had not been advised of the existence of this potential "unsafe condition"!


They were asked to accommodate the lack of pitot probes Thales AA and ensure accountability of the "unsafe condition" by applying a checklist when it is defined that an "unsafe condition" may lead the pilots to excessive workload which does not allow them to perform their tasks accurately or completely!”

* http://henrimarnetcornus.20minutes-b.../102670700.pdf

** http://henrimarnetcornus.20minutes-b...1239787795.pdf

ChristiaanJ, there is another little detail you may not have thought of. Communications does not have to be via satellite if an alternate link can be established. That greatly reduces the demand on satellite communications. An extended cell phone technology could handle large cells from the air. If the plane is too low and falls out of the large dedicated to aircraft cells you use regular cell phone towers. Only if they fail you fall back on satellite linkages.

Of course, who is going to pay for the development of the large communications cells when no worldwide frequency allocation exists for this purpose; and, getting it will likely be a more than one decade process given how often the international frequency allocation body meets.

It's all an expensive non-starter. It can be done. It'd he fun doing it. It'd be full of interesting challenges. And it would never "pay for itself".

{^_^}

Wake up Mods!
 

It's time this thread was locked, a new one opened for AF447 discussion only, and another in Jet Blast for impossible crackpot ideas.

There's always this one, SallyAnn.

http://www.pprune.org/tech-log/376433-af447-230.html

It seems to be the major repository of both crackedpottery and some good analysis. Let those with "ideas" for solutions mine that thread to see if they've presented anything new. For example, satellite FDR/CVR is very much "not a new idea." So far a cockpit video recorder linked to base by a satellite connection has not been proposed. So it's not THAT far "out there."

{o.o}

Sallyann1234,

There isn't really much in the way of news on the subject anyway, at the moment, so why not let it run, as most of those interested are monitoring it?

So far I've seen almost no comments on SPA83s posts, although he's pretty well succeeded in convincing me as to the "probable cause".

CJ