CVR is found.
BREAKING NEWS: CVR Black Box Sriwijaya Air SJ 182 Ditemukan (kompas.com)
Indonesia recovers cockpit voice recorder of crashed Sriwijaya Air jet (yahoo.com)

Fantastic news

Very good news indeed, but let's wait to see if it can be exploited...
In any case congrats to the salvage team !

Fantastic search team, respect.

That's absolutely brilliant, well done to every one involved.

I was certain it was gone for good.

737-500 missing in Indonesia
 

Just a thought but but why not have duplicate CVR’s and FDR’s on aircraft in different locations?

Good idea Brian. Better than impractical idea of broadcasting it for ground storage.

More expensive , perhaps, but not impractical.

Kudos to the diving team working in those awful conditions and probably finding non airframe pieces too.

ktcanuck

"More expensive, perhaps, but not impractical."

Or, if all else fails, just attach the ULBs more securely to the memory modules ...

Still all-eggs-in-one-basket rather than redundancy - but multiple would carry weight penalties and temptation of making them less crash-survivable and risking all of them not surviving, tradeoff as always.

But with what you say, is there a reason the ULBs have to be separate and "attached" (or detachable...), why not integrated?

I have vague idea this may be to do with ULBs having to be regularly replaced as the batteries die, but these days with solid state memory and near-field wireless power and data interfaces is it now possible to do ULB / CSMU in a single easily replaceable module?

How would you suggest to trigger an integrated ULB? As the current models are set off by immersion into water, that would not work quite well if encased.

Great news - as above I'm sure its been a very tough job. Physically unpleasant .

There's a tendency to slag off the Indonesians on here but they really stuck to this task

Ref post 728. When you see what Airlines are doing to save money / weight / fuel costs I cannot see them duplicating anything.

https://www.aerotime.aero/23619-17-t...irlines-change

Modern FDR’s and CVR’s monitor so many things the How it happened & What caused it can now more easily be found. Until such times as Cockpit Video Recorders are installed the Why can sometimes only be conjecture.

Why is that impractical?

Data rates these days are far in excess of what is required, and storage costs absolute peanuts.

Not only could it be used for accident investigation but in these days of big data crunching with machine learning and artificial intelligence, that data could be used for failure prediction. That sort of technology is already used in other operational technologies. In the long run it could pay for itself if predictive maintenance prevents "unplanned events".

And imagine if all of those sequences of events that lead to aircraft loss are combined and analyzed, and pilots could be warned that a seemingly harmless alarm taken in context of other data indicated that a catastrophic failure was in the offing?

Added: And automatically gave him a solution?

https://analyticsindiamag.com/how-to...-data-science/

The idea of storing CVR and FDR data outside the aircraft has been researched and discussed by all the major international aviation agencies in the recent past (especially ICAO, of course). The notion gained momemtum because of AF447 (recorders eventually located and retrieved) then again because of MH370. Of course the idea has benefits but it also has some significant hurdles.

Pretty likely that they are attached feebly by intent in order to protect the memory module. Musk's Starlink will make it all moot soon anyway. There will be thousands of satellites (about 1,500 today?) in low earth orbit serving as communications satellites. Price today is £500 setup + £100 a month for unlimited data.

At present only fixed base stations are supported but that is changing. Not freely available I don't think, still under some sort of test regime with limited coverage.

The commoditization of beam steering, flat plate, phased array antennas may turn out to be Elon Musk's greatest legacy. You probably didn't know you wanted one.

There was someone on another forum they other day who said something like - well it's like a moveable dish but it doesn't move, but I can't say any more 'cos its a military secret. You can now buy them for £500.:)

https://en.wikipedia.org/wiki/Starlink

It's been discussed to have the CVR and the FDR 'mirror' each other - i.e. have both record the same data. As memory has gotten smaller and cheaper, it's now practical.
It would require some up-front expenditures to add the capabilities, and some wiring changes to the aircraft, but it's certainly 'doable'...

The biggest drawback to 'real time' data download is recurring costs - the satellite band width doesn't come cheap.

Realbabilu, thank you for posting the announcement about discovery of the CSMU.

A hearty "Atta boy" to the KNKT and everyone on Team Indonesia! The difficult took a few days, the impossible took a little longer.

Here is a close-up of the CSMU shortly after it was located, from video accompanying this Kompas article:

KNKT: CVR Sriwijaya SJ 182 Found After One and a Half Months of Searching
Author Sania Mashabi | Editor Diamanty Meiliana
Kompas.com - 03/31/2021, 13:04

https://cimg2.ibsrv.net/gimg/pprune....b43d965599.png

Reports indicate that the CSMU was found by employing the services of the "King Arthur 8", a class of vessel known as a "Trailing Sucker Hopper Dredge" or TSHD. Here is an image of the ship

Talk about using a huge cannon to swat a fly !! That is in no way a criticism, just an observation about the SCALE of the vessel, relative to the approximately 100mm tall by 100mm OD dimensions of the CSMU.

To get some idea, watch the following YT video:

"Dredging: Trailing Suction Hopper Dredgers - How does it work?"

Considering the large volume of material a TSHD vessel is designed to rapidly process, it is hard to imagine how searchers pored through the contents deposited in the ship's hold.

I doubt there was much that could be done to throttle back the rate of collection fro the seabed - there must be a minimum suction required for effective transport up the pipes. Placing screens over the outlets seems improbable because of the frequency of blockage by crash debris and other solid material.

Perhaps industrial sized mesh strainers were suspended in the hold below the outlets of the pipes.

Even something crude like dumpsters with numerous holes smaller than 100mm x 100mm torched into the sides for drainage would have sufficed.

Whatever they did, congratulations to them ..

The article above says that they were only searching a 90m x 90m square, to a depth of 1m ... so about 50cm thick over the bottom of the holds once they had the material on the ship, perhaps.

The CVR was found on the night of the last day of searching.

Does anyone think we will ever see a transcript?

Yes, for the major accidents in Indonesia for the past 8 or 10 years, in the English version of the reports, KNKT has included a detailed translation of the CVR transcript (timestamps, who spoke the words, etc.). Some phrases or sentences are summarized (i.e. "PA announcement made to passengers") and some direct quotes / detailed info is included for what KNKT considers more relevant to cause.

Just FYI: Some national investigative agencies routinely include full transcripts as appendices to their reports and some do not. See Annex 13, Sections 5.12 and 5.12.1 for the ICAO position and guidance.

The agencies can discuss it all they like, but the practicality will come from the manufacturers, the likes of Honeywell etc. You can't require what isn't available on the market.

Actually, you need both, the technical feasibility and the regulatory framework. Thales or Honeywell can design marvellous things, they will remain very expensive prototypes in the absence of regulatory material, to ensure their marketability. Requirement and Advisory Material will provide some minimum specs as well as ensure standardisation (a vital aspect for CVR/DFDR), and will typically mandate the equipment or give it some kind of credit (replacing older equipment or allowing further operational possibilities) so that it can be access the market.

As an example, EASA has published various NPA (Notice of Proposed Amendment) and is funding research on the topic, just Google it.

WHBM

Not to jump on the wagon, hello ADS-B OUT

armagnac2010

Most commercial aircraft are still flying around with the same model of CVR and FDR they were delivered with decades earlier. It is rare that updated recorders are fitted to existing aircraft, and it's very expensive to do so. To mandate retrofit would require an Airworthiness Directive - and those have to go through a cost/benefit analysis to justify the expense, and it's very difficult to quantify the financial benefit of improved data recorders.
There are still commercial airliners out there flying around with the old foil style recorder - in fact about ten years ago I was investigating crash FDR data from an analog foil recorder (the only engine parameter was EPR - and that of questionable accuracy).

mnttech

"Not to jump on the wagon, hello ADS-B OUT"

ADS-B-Out provides a minute fraction of the bandwidth that would be required for real-time FDR/CVR streaming.

ADS-B-Out is a welcome advance in aviation and will surely provide benefit, but, as Dave suggests, there are still BIG issues re bandwidth when we are talking about real-time streaming of all CVR-FDR datapoints for all commercial aircraft.
Just as importantly (IMO) is the ongoing discussion of the intent of CVRs and FDRs. Adequately protecting the privacy of the information (especially from CVR) and ensuring it is used solely for its intended purpose (official accident investigation to ascertain cause and contributors) is problematic with streaming.

A single Starlink channels can give 200 mb/sec and the average number of planes in the sky over the last few years is 10000 at any one time giving about 20000 bits per second from one channel assuming they are all equipped. This is from one channel and there are many available.
Bandwidth is not really an issue

, but costs are.
Over continental airspace bandwidth will definitively be if using conventional Comms links, which are already saturated with mode S data., outside VHF coverage you will need satellite comms.
Satellite comms are still very expensive because all privately owned or run , and airlines are rejecting the idea of mandating it. Been discussed already for years within ICAO.
And as Dave pointed out already, ADS B-Out is not made at all for this.

How much is "very expensive"? If they all have to do it there is no competitive disadvantage

They're happy enough installing the internet for passengers

My point was that once the ADS-B mandate (at least in the US) was put in place, technology expanded to meet that requirement and more. I agree that the bandwidth is not there, nor was I saying ADS-B is the answer, I was simply showing that the FAA mandate (10 years prior to 1/1/2020) drove industry to come up with solutions.

Would it be feasible for ADS-B to output useful parameters at a relatively low refresh rate under normal circumstances, but in a recognised abnormal situation where certain parameters are exceeded, for the system to increase the amount and/or frequency of data transmission? Not to replace FDR data, but to augment it and allow investigators access to some early data more quickly.

How about this. The FDR and CVR keep ticking away as usual but, and here's the rub, under some circumstances the data they capture is also transmitted. The hard part would be defining those circumstances but they could be as simple as any slight deviation from expected parameters. Bandwidth is not the issue since engine data is transmitted and passengers get internet access. Far from perfect since some of the context might be missing in the case of total loss

Ok , this was discussed here many times before over the last decades, and was even the subject of an ICAO working group in the past . .In a nutshell : technically feasible, yes on paper , but the cots to set this up are enormous and the benefits did not match the expenses.
It is not only the transmission costs of using the Satcom , the amount of data to uplink and collect but also the storage infrastructure needed and the eventual retrofit to existing fleets all of it to be borne by the aircraft operators.
They all have said a big NO at the time . And since decisions like these are always done in consultation with airlines, no chance to mandate this against the will of the airlines I would say.
Yes because it is a commercial business decision, they believe if helps them getting more valuable business pax , and they sell the time at a premium .

Int'l Aviation Agency
 

Reference was made to past discussions within ICAO -- so here, a question about that.

Where grizzled (738) referred to past discussions and noted specifically AF447 & MH370, were these discussions in the context of the subject, Global Flight Tracking? I was present (observer status) for part of those. Were there other discussions on the subject at, for example, the Air Navigation Commission?

Where ATC Watcher (753) also mentioned past discussions, were those separate from the Global Flight Tracking focus? [Plus 758, working group.....added]

No doubt offices of Permanent Reps have little difficulty getting into ICAO records quickly, but a cheap-seats SLF/lawyer gets by with a little help from my friends (please).

Global Flight Tracking and FDR/CVR streaming are two separate issues.

You beat me to it, Dave.

They are somewhat related, but only in the sense that global flight tracking became a hot topic issue due to M370 (i.e. if we were far more advanced with global flight tracking we would have fewer reasons -- note I didn't say "no reason" -- to consider FDR / CVR streaming).

OTHER than in the event of an accident, under what circumstances and how frequently is FDR/CVR data accessed? Not often? Then real time remote transmission/storage (and its technical challenges, complexity, fallibility, associated bureaucracy, costs and other hurdles) seems like an unduly expansive response to the problem at hand.

Like most black boxes, SJ-182`s FDR was retrieved and read efficiently. It is SJ-182`s CVR which proved difficult to locate (in part because the ULB separated from the CSMU sub-assy of the FA2100-style recorder.)

One solution: enhance legacy FDR & CVR devices to be more reliably easy to find.

K.I.S.