Thanks for these clarifications.

As stated, several factors weigh heavily against satcom-based streaming - yet the idea gets brought up now and then. So let's say there are research interests in the way ICAO considers problems and possible solutions with somewhat complex cost-benefit assessments, or in developing new segments for international air law curricula, or in preparing background material to be used by one of the sector organizations or specialized government agencies.

Rather than taking any side on the merits this SLF/attorney is only looking at how digging into the record of ICAO's review of the issues might be pursued, below the easy superficial (official) record that's readily accessed. (Such is the dullness of legal and policy research....)

"As stated, several factors weigh heavily against satcom-based streaming"

1) Cost of streaming, prohibitive?
2) Current regs:

• ICAO Annex 6, Operation of Aircraft, Vol I, Attachment D and Vol III

Nuff said!

Real time streaming of cvr/fdr
 

One problem that would seem to exist in using existing satellite technology, is that it doesn’t work well during unusual attitudes and/or dynamic flight changes. You can observe this yourself when you lose tv signals and/or internet during an ordinary turn.

Surely you’d want to solve that problem as well, since some of the most useful information will happen outside of straight and level flight

Can I bring this post back into the discussion? To an outsider, it seems like a highly proportionate measure, and it seems that not infrequently one recorder is found, but not the other. Any non-obvious problems?

FlightlessParrot

"it seems that not infrequently one recorder is found, but not the other."

I don't think the facts support that assertion.

Well the main goal of airlines is to make a profit no? . I am no fan of IATA and beancounters but here I have to admit they have a point.
Mandating changes without looking at the costs to the end user is a thing of the past. . We are now working with consensus and while nobody has a veto, everybody listen to the other side, and when what they say make sense , the group concur. This is how we now work in international technical panels , working groups and other bodies where regulations are being discussed.

The airlines have been fooled before in spending big $ on equipment that never been used as planned , think MLS supposed to replace completely ILS by the year 2000 to take only one of many examples..
To end this discussion :
a) in our particular Indonesian case here , the FDR and CVR have now been retrieved and FDR/CVR data streaming would only have speed up the results but not changed the outcome
b) Data streaming in real time will probably come naturally when 5G internet will be available via satellite at a fraction of the cost it is today, and one or 2 large countries would agree to build , run and secure a massive data center to store all this, and then agree not to used the data collected inside themselves, and that data will only be accessible to recognized Foreign National Accident Investigation Boards..
This is probably not going to happen in my lifetime.

DaveReidUK

Agreed, but there have been cases where only one had usable data (Lauda 767 the DFDR was unusable although the CVR was - much of what we know about the actual T/R deployment came from NVM from the event engine FADEC which was rather miraculously intact).
Less of a concern with the newer, more durable recorders, but it can still happen.

568

It's not simply a matter of "how much" does it cost. Ask also, how much does it cost not to have it?

If an airplane is down and lost, how much does the search cost? How much does it cost to suffer a repeat event when the cause of the original accident is unknown and unmitigated? How much does it cost fruitlessly looking a variety of possible causes that could be ruled out with recorded data, if available? I'm not presupposing the balance here, but the way technology is advancing, the costs becoming easier to afford.

Not to beat an oft repeated saying, but if you think safety is expensive try an accident. Lives could be in the balance.

A new user hit the nail on the head.

"One problem that would seem to exist in using existing satellite technology, is that it doesn’t work well during unusual attitudes and/or dynamic flight changes. You can observe this yourself when you lose tv signals and/or internet during an ordinary turn."

Pretty hard to connect to a satellite If the plane is in a steep descent, bank or inverted.

The last moments may not transmit well, true, but things start to go bad before that. And that few good data points may help locate a downed aircraft

DaveReidUK

You would know better than me. OTOH, from a cost-benefit point of view, once you've found one you can stop looking for the other, which would have saved a fair bit of money here. I guess it depends how much it would cost to do the mirroring: on the ground, almost trivial: in aviation, ?$?$

DaveReidUK

Not including this accident, Wiki lists 38 accidents where at least one recorder was either unrecovered or unusable. Of these 38:

15 - Aircraft wreckage located but no recorders found
4 - Aircraft wreckage located, FDR found, CVR not found
2 - Aircraft wreckage located, FDR not found, CVR found
10 - Aircraft wreckage located, both recorders located, at least one was unusable due to damage (in some cases, potentially only one installed)
7 - Aircraft wreckage never located, or other nefarious dealings involving corrupt governments.

https://en.wikipedia.org/wiki/List_o...ight_recorders

GlobalNav

I am in agreement with you that is why I placed a question mark against the first bullet point as most operators don't want to be a part of the additional financial costs.
SAT comms are also an issue with some coverage at certain latitudes being very spotty.

slacktide

Thank you for making my point - in 55+ years, there have been 6 events where mirroring of the FDR/CVR contents might have made a difference to the investigation, and a further 10 where the potential benefit is more questionable.

Good luck making an economic case for data streaming from that.

Last point on the cost comparison between CVR/FDR search and data streaming : Search costs are taken up either by States or by insurance companies, not all by the airline. and is around one in a lifetime per airline.
Data streaming costs will be 24/7 paid by the airlines .

And once again when mentioning MH370 , please do not mix Global Tracking and data streaming . .
Global tracking is easy and relatively cheap, a lat/long /ALT short message every minute or so using existing the Iridium network is around 50$/hour for the transmission part and every Airline OPS can have a PC to store the data of its own aircraft,. Some small airlines operating in hostile areas already use it . Data Steaming is very complex, especially rulemaking wise, and very expensive.

DaveReidUK

Not many cases, but the point about the proposal tdracer mentioned is that there would be NO data streaming. This is about copying the data from one recorder to the memory of the other one, and vice versa. Mirroring. All contained within the aircraft. NO streaming. No connection with the outside world. No data transmission costs.

As you say, max. of 16 cases where it might have made a difference, and the one-time cost of designing and implementing the entirely-within-the-aircraft-without-any-data-streaming mirroring solution might well exceed any quantifiable benefits, but I'm surprised it is not discussed more as an ALTERNATIVE to data streaming, which it is NOT.

Did we get any update about the condition of the recovered CVR ? Is it usable !?

Latest report indicates the CVR has been read and includes the last minutes of flight. Good news, but whether it explains the Pilot's actions remains to be seen from further analysis and alignment with the data recorder.

https://www.channelnewsasia.com/news...oaded-14607864

Hunbet

Not just that, we can't even get SATVOICE to work reliably, in that it fails at a network level far too often. This is why HF carriage is still mandatory on the NAT for example. CPDLC and SATVOICE are effectively the same system. Not to mention a continuous stream for every aircraft would be unjustifiably expensive for the industry.