How long can an airplane's black box last before degrading due to natural causes? Forever?
More precisely: Assuming the black box survived the crash intact, and assuming that it will be located between 1 and 1,000 years from now, is it guaranteed we'll be able to pull useful data off of it? I wonder what the engineering process is like for one of those things...
EDIT: Apparently, the key component of a flight data recorder is its Crash Survivable Memory Unit (CSMU). This manufacturer claims a lifetime of 30 years: http://www2.l-3com.com/edi/crashsurvivable.htm
The CSMU is rated for 30 days of saltwater submersion. Obviously, the only scenario it'll be submerged in saltwater outside of the lab is if the recorder's housing cracks. So, assuming the recorder survived the crash, I wonder which would degrade first: The housing or the memory unit? If the housing fails, then the memory unit is toast, since it's only rated for 30 days of saltwater. The manufacturer guarantees the CSMU will last 30 years, but it seems like as long as it stays dry then it might last for much longer. I wonder how long it's theoretically possible to extract useful data off of it assuming it stays dry.
Regarding the saltwater submersion: I think your scenario is not correct.
The whole orange thing displayed on that page will be have to be in contact with saltwater. Otherwise the ultrasound pinger (the cylindrical aluminum thing) doesn't even start emitting pulses... (conductive liquid on sensor pads is how they normally detect that they have to turn on)
The "lifetime" of 30 years, or 17'000 operating hours is for how long it's certified to run in a non-crashing normally operated airplane.
The 30 days in saltwater probably means that no moisture will have crept in through the covers and seals: You can rinse the power-connector and Ethernet with deionized water, let it dry, connect to the computer and read it out.
{A assume that these devices are read out much more often in cases of non-fatal crashes where they weren't subject to the elements that much.}
The block diagram they show has an separate component for the nonvolatile memory, in a "Crash Survivable Housing". Also I assume that the company at least has some plan for accessing data given a complete disintegration of the memory module, probably by contacting single flash chips.
EDIT/ADDED:
http://edition.cnn.com/2014/04/23/world/asia/malaysia-370-bl... says: """A lot of our work is with undamaged recorders and it's very easy to download them, much as you would a USB memory stick,(...)""" and """One of the most challenging scenarios is when the board itself is damaged: "We could take each individual chip off the circuit board, read those out individually, and then with the help of the manufacturer, piece all that information together,"""
It may be practically impossible to find with today's technology, but if the next 100 years of progress is anything like the last, there'll be large scale hobbyist exploration of the sea floor.
The surface area of the world is 510,072,000 km², 70% of which is ocean. That's absurdly large if you're exploring it with 8 figure submarines, but not at all if you're exploring it with the 2114 equivalents of 2014 $100 aerial drones, especially given a larger, wealthier society with more leisure time.
And work aside, think about living life without modern conveniences such as washing machines, dishwashers, fridges and freezers, vacuum cleaners, electric/gas ovens, microwaves, kettles and air conditioning.
Running a household a hundred years ago (or fifty years ago depending on wealth) would have been a much more time consuming affair for a significantly lower standard of living. I expect that trend to continue.
Well, if we assume that we're looking for an aircraft with a few large pieces, and it's only a matter of time before cheap autonomous submersible vehicles become common, then it probably will be found some day. Might take a decade or two, but at some point we'll extensively map the rest of the oceans.
I guess the question becomes who is going to pay for these robots? They'll need the ability to dive deep, explore, and then come up for "air" (a recharge). Those aren't going to come cheap. Perhaps the undersea cable conglomerates but it's hard to imagine the use-case for mapping inaccessible areas.
Extend past a few decades. A post-scarcity society would almost certainly search every square metre of the ocean for this airplane, assuming humans are still making decisions.
Extend past a few decades and you're beyond the lifetime of the black box, sadly. It would be really frustrating to finally find the plane only to never actually find out what happened.
Obviously there will be huge differences between different parts of the huge oceans, but as a single data point: I've seen photographs taken of the Titanic wreck, where still single shoes can be seen laying uncovered on the ocean floor in the vincinity...
I don't like that comparison, because when I think of San Francisco I think of lots of buildings complicating the search by requiring you to go floor to floor, but the comparison is just looking at land area, so it's easier than it sounds (but still really tough).
For mine, the topographical variations of the ocean floor (combined with the fact they're largely unknown, at least to any level of detail) are probably similar to a floor-by-floor search, but you're right than basing it just on area makes for flawed comparisons.
I've looked over the calculations and I think they are making a very shaky assumption.
The BFO-bias is calibrated why the plane is on the ground, and they assume it is constant over time, temperature, air-pressure etc.
This is probably not a valid assumption under any circumstances, given that there is no atomic frequency standard involved and probably not even an OCXO.
Given that MH370's inmarsat-terminal was powered off for something like an hour during which environmentals may have been extreme and even outside specs, the assumption of a constant bias certainly have no legs to stand on in this particular case.
If the terminal used something like a 1PPM milspec 100MHz TCXO, I would guess is that the BFO bias could be anywhere between 0 and 300Hz after the 18:25Z logon.
I have not tried to replicate their trajectory estimation using this uncertainty, but by eye-balling it, it vastly increases the uncertainty of the resulting location.
The validations they have done using other aircrafts involves BFO's running in nominal mode throughout, and therefore provides no information about how the BFO bias might change after power on/off or extreme environmentals.
Hopefully this is not too much off-topic but one of my best school friends was murdered on flight UTA772 in 1989 (he wasn't the target of the bomb, just one of the consequences). Seven years later his mother forwarded me his camera that had been discovered in the wreckage. It was pretty beaten up but it seemed to me that there was still a film in it. I managed to ease the film out undamaged but by now it was way past its process by date and had been lying in the desert for a while so I wasn't at all optimistic about the latent image. With technical help from Kodak and a very good lab we eventually managed to get some usable images off the film. To my amazement there was, when greatly enlarged, one picture of him. Grainy and colour distorted but definitely him. This picture was some source of comfort to his family as his body was never discovered which, as is always the way with these things, adds immensely to the distress.
So on MH370 there might be other useful sources of data in the wreckage apart from the FDR.
Often the wreckage itself tells a big story, assuming most of it can be found in a debris field. It can certainly say whether it broke up in the air or hit the water intact, for a start. Then you can look for bomb explosions or other types of explosions. Even the location of the wreckage would fill in large parts of the story which is currently unknown.
The interesting thing to me is how much money,collectively, are we (and the term is loose because I don't know who is paying) prepared to spend to find this wreck?
The other way of looking at the expense is "How much are we willing to pay to prevent another such disaster?" (assuming it wasn't a conspiracy of the pilot).
Battery life ... I think the 30 days figure is the minimum spec, but in reality it can go on longer, depending on age of battery, i.e. when it was last replaced.
It says that the shelf-life of the battery is only 6 years (which probably means that it will lose some significant part of its total capacity over 6 years). Having a device operate for "decades" surely will need different power sources. (even if cutting down on average power consumption drasticly, by only emitting one ping per week or so...)
I think it just comes down to this being an extremely conservative field, and these systems get implemented based on past experience.
Before MH370, basically nobody anticipated having to search for a commercial airliner with no idea where it went besides that it was vaguely somewhere within a continent-sized patch of ocean. The 30-day period was considered to be enough. Even with MH370 it could have been enough if the Malaysian authorities hadn't been so unbelievably incompetent as to waste an entire week searching the wrong ocean when they had data which definitively showed it went elsewhere.
It's similar to the question of why airliners don't periodically report their position using GPS and satellite data, since they basically have the capability already. The answer is that nobody really thought it was necessary, since tracking airplanes through other means worked fine right up until it didn't.
Here's a TL:DR for you: "The analysis presented in this paper indicates that
MH370 changed course shortly after it passed the Northern tip of Sumatra and
travelled in a southerly direction until it ran out of fuel in the southern Indian Ocean
west of Australia. A potential flight path has been reconstructed that is consistent with
the satellite data, indicating a last contact location of 34·7°S and 93·0°E, but it is
stressed that the sensitivity of the reconstructed flight path to frequency errors is such
that there remains significant uncertainty in the final location."
That doesn't really explain the dramatic change in course. To essentially do a 180 degree turn and fly the opposite direction there must have been some sort of pilot input into the navigational system.
It's amazing how stupid people can be when oxygen deprived. My dad had decompression chamber training with the Aussie air force. They gave everyone a clipboard questionnaire with some simple puzzles. In oxygen-deprived state most people were writing off the edge of the clipboard and on their legs.
True, but it doesn't really fit with disabling the transponders and changing course moments after communicating normally with ATC.
Any theory other than deliberate action seems to rely on several major coincidences. Not strictly impossible, but definitely not the most plausible option.
Pilot suicide has happened before. Some of what happened, such as possibly ascending to a high altitude early in the flight is consistent with a pilot incapacitating the passengers and the rest of the crew, and maybe himself, too, by outlasting the emergency oxygen. If the pilot intended to conceal evidence of what happened, setting a course to a very remote part of the ocean would be consistent with that goal.
More precisely: Assuming the black box survived the crash intact, and assuming that it will be located between 1 and 1,000 years from now, is it guaranteed we'll be able to pull useful data off of it? I wonder what the engineering process is like for one of those things...
EDIT: Apparently, the key component of a flight data recorder is its Crash Survivable Memory Unit (CSMU). This manufacturer claims a lifetime of 30 years: http://www2.l-3com.com/edi/crashsurvivable.htm
More info: https://www.aea.net/AvionicsNews/ANArchives/TechSpeakMar10.p...
The CSMU is rated for 30 days of saltwater submersion. Obviously, the only scenario it'll be submerged in saltwater outside of the lab is if the recorder's housing cracks. So, assuming the recorder survived the crash, I wonder which would degrade first: The housing or the memory unit? If the housing fails, then the memory unit is toast, since it's only rated for 30 days of saltwater. The manufacturer guarantees the CSMU will last 30 years, but it seems like as long as it stays dry then it might last for much longer. I wonder how long it's theoretically possible to extract useful data off of it assuming it stays dry.