fgrieu

Yeah, once you've exposed the silicon chip inside the package, in effect you have what looks like a crater burned through the plastic (epoxy). You can see the entire square silicon chip, with the ends of the lead frame and wires joining to it.

Then once it's all nice and clean, you add a drop of liquid crystal (similar to the stuff inside your laptop screen or LCD watch) into the crater to make a thin coating. Then you shine a focussed scanning beam of polarised laser light over at the right angle and the reflection is either polarised or not- depending in the charge of the bit. In some cases you need to set up a weak electric field across the chip, just like inside a LCD display. You then use a linear CCD pickup with a polarising filter to read the data out. The trickiest bit is aligning everything so you only scan one row at a time. Kinda like an old tv set, line by line. Doesn't take long once it's set up. Used a lot for CPU and ASIC development and fault analysis by big silicon chip companies.

You may have seen video footage of microprocessors with their lids removed ticking away... Also done with a drop of liquid crystal and a ploarising filter.

Here's a pic showing how you disolve away the plastic to expose the chip.
***link removed***
There are quite a few youtube videos showing the circuits working under the microscope too.

The simple questions I asked have still not been answered. Instead I was subject to the usual “I’ve flown in and out of many places in Indonesia”, “You don’t know this place”, and even comments about vehicle drivers in Indonesia.

Though born and brought up in “the West”, and spending most of my aviation career there, I spent much of the past 15 years living and working in Indonesia -- including working directly with the NTSC on several accident investigation files.

As with other recent posters – and long-time ppruners – I’m not optimistic regarding the ability of this site to be a “professional” forum. That objective is difficult to achieve these days and pprune is an example of sites that are trying to figure out how to accomplish that. In the meantime, I shall lurk – with interest and curiosity – but I will be more reticent to post. I thoroughly enjoy discussion and debate but I have no interest in the abrupt and confrontational tendencies of internet-based communications – rather than nuanced and respectful interactions, as one would encounter in a pub, coffee shop, or even with the person in the adjacent seat on an aircraft.

grizzled

I don’t believe in conspiracy theories.
I’m talking about basic competence.
Indonesia is the only place I’ve operated into were I decided to just assume bad thing were going to happen.
It was safer that way.
ATC has improved a little over the years but can still lead you down a rabbit hole.
Weather forecasting is still woeful.
Standards at LCC ?
You tell me if you work there.

The Indonesian National Transport Safety Committee (KNKT) have issued a statement saying that all the data from the FDR has been successfully downloaded.

They have additionally stated that there are 330 parameters in the device, and "all are in good condition". The search for the CVR is continuing.

https://www.reuters.com/article/indo...-idUSL1N2JQ0BN

Sadly they are one of the most experienced outfits at crash investigations these days...............

Good to hear that they could read out the FDR. If is was a technical problem and they have the last 40 seconds recorded, they should have already a decent idea what happened. The CVR might not reveal much more in this case with the two poor folks in row 0 fighting for their lives.

Engine failure ruled out
 

KNKT confirmed at yesterday's press conference that both engines were operating up to the point of impact.

George Glass

George,
Re your points above:
1 I have no idea why you mention conspiracy theories.
2 Pease tell me what "basic competence" has to do with your original comment saying it would be "very convenient" if the CVR was unreadable (which I asked you to clarify).
3 I certainly agree that your mitigation strategy re assuming bad things would happen is a good idea with respect to many aspects of civil aviation in Indonesia. As it is with other places I have lived and worked (most notably some parts of West Africa, and some other parts of Asia).
4 Re ATC in Indonesia: Agree completely with your comments.
5 Agree completely re Wx forecasting (and I would add Wx briefings as well).
6 Agree completely re standards at most Indonesia LCC's (meaning operational and safety standards -- and culture) being an issue. I would add airports (especially smaller regional airports into that mix).
7 The NTSC / KNKT is one of the better aviation investigative agencies in the region -- as attested by other international agencies and by the quality and thoroughness of their reports for the past few years. In that role they are certainly doing more than the regulator to address and influence the safety culture, and corruption issues, in Indonesia (baby steps).

The FDR will be enough to tell the whole story. There would be no reason to "hide" any CVR.

Without the CVR the MSR990 crash investigation by the NTSB in 1999 would not have come to this:

"Two weeks after the crash, the NTSB proposed handing the investigation over to the FBI as the evidence they had collected suggested that a criminal act had taken place, and that the crash was intentional rather than accidental. The Egyptian authorities refused to accept the proposal to hand the investigation over to the FBI, so the NTSB solely continued the investigation." (Wikipedia)

Well, to tell the technical side of the story to be more precise then. You can monitor all control inputs.
Don't know about the legal side of things. But this is not what accident investigators do.

Less Hair

It depends on the cause - if it's an obvious technical issue, then I agree, the CVR is unlikely to add much. OTOH, if - as donotdespisethesnake suggests - it's a baffling LOC without any major aircraft failure, the CVR will likely be key. How much would we understand about that Air Asia A320 if we didn't have the CVR to know what the pilots were doing?

Air Data/PFD failure, possibly switched to LHS/RHS in order to dispatch

The commander of the Indonesian Navy's First Fleet Command, Cdr Abdul Rasyid has reported in a media interview, that divers have found parts of the CVR.

Cdr Rasyid stated, "We have found the casing, the beacon and the CVR batteries. We need to search for the memory unit. We hope it will be not far from them" (the other parts found).

This is not sounding good, if the CVR casing has burst open with the impact. Not only is it making the finding of the memory unit more difficult, there is a high possibility the memory unit may be damaged, to the extent it doesn't yield useful information.

Quick question: Can the pilots pull the CVR CB during flight? ..... or even before take-off?

cb is located on panel P18, shoulder height. (Behind captain's seat). Easily reachable. On the triple I can't pull it easily(unreachable in EE compartment) but on the 737 you can. Why?

It depends what the Navy spokesman meant by the "CVR casing".

It's not clear which vendor's CVR the aircraft was equipped with (flight recorders are BFE, specified by the customer). It will almost certainly have been either an AlliedSignal/Honeywell or Fairchild/L3 unit. Both share the same basic construction: a chassis, onto which is mounted (i) a box with circuit boards and (ii) the memory module. The latter is encased in a hard shell, typically stainless steel or titanium, purpose-designed not to fracture in an impact. The ULB is mounted on the memory module, since of course that's the bit that it's most important to find.

So it's not immediately obvious which of the above the spokesman was referring to when he referred to the "casing". I wouldn't rule out finding the memory module shell with its content protected and intact, as it is designed to do.

Edit: Found a video of the recovered CVR chassis from yesterday, which confirms it's an L3 (originally Fairchild) FA2100. It is indeed mangled, with all the electronics visible - but that's the non-crash-survivable part of the CVR and its condition has no bearing on the likely state, when found, of the memory module:

fox niner

I my company, in the event of an accident or serious incident we are expected to pull the CB’s for the CVR and FDR after shutdown. This is to stop the useful data being over-recorded while the jet sits on ground power. The 777 probably has far bigger memory banks so this procedure is not needed and the 777 CB’s can be berried.

over the years there have been many cases of pilots allowing the CVR to continue to record after landing - this overwrote all the information recorded during an "incident" on early small capacity recorders. This was especially the case for the CVR - even the FAA started to get suspicious.

In this type of accident the FDR is far more important than the CVR and will most probably give results as to what happened in the last minute of the flight. , If you take previous similar accidents , like the Air Sweden CRJ as an example , the CVR alone would not have explained what happened. The early info released that both engines were giving power at impact eliminates already a few scenarios.