Sarcs #954 –"A quick Sunday afternoon Google search for unrecovered black-boxes turns up a list total of twelve (I guess soon to be 13) from 1965. Some of those cases it wasn’t from a lack of trying to retrieve them. And sometimes they were totally or partially destroyed or lost".
Good catch Sarcs, the rest of the world seems to be able to make an effort, mostly successful it seems to recover "hard" evidence. I couldn't find a copy of the ATSB policy or investigators manual, but the Canadians and the NTSB seem to have no problem making public theirs :-
NTSB Investigators Manual.
From page 156 (PDF) – there are guidelines for OBR which make no bones about how important the recovery of the CVR etc. is to their investigations, most refreshing attitude. Now I know the manual is 'dry' reading, but there is some really good stuff tucked away in the 'procedural' parts worth cherry picking, just for reference.
I also found this from an interview with NTSB 's James Cash:-
Would the thing even spit out anything useful after four months in salt water? While we can’t answer the first question, we talked to James Cash, chief of the vehicle recorders division at the National Transportation Safety Board, to get some answers to the latter two.
Cash also says it is absolutely possible that a plane’s black box would still be in working condition after four months on the ocean floor. The current record goes to a recorder that, after nine years at the bottom of the Mediterranean, was perfectly fine. “The water, in general, doesn’t hurt them at all,” Cash says. “It’s the air that hurts them once they’ve been wet. It starts the corrosion and rust process.”
Once the black box is found at a wreckage site, it’s transferred to the lab in a water-filled cooler so the data can be retrieved and copied right away. So although we know Widmore has the black box, we’re still not entirely sure he has a copy of the data that it recorded.
According to Cash, it only takes a few days before the leads on the SSD start to corrode. The exact time that needs to pass before the data’s not retrievable is unknown. “I would say a couple of days,” Cash says, agreeing that a month out of water is probably sufficient time to render the black box useless. “We try not to experiment with that so I don’t know if I have a good answer. But if you let it dry out for quite a period of time, it’s going to make [data] much more difficult to recover.”
Cash interview.