The Search for MH 370 (Inmarsat)

http://journals.cambridge.org/download.php?file=%2FNAV%2FS037346331400068Xa.pdf&code=99d6daf127f9c88ca22b52fd9ff5084a

by wglb • 12 years ago

86 42 12 years ago

journals.cambridge.org

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

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.

baudehlo3• 12 years ago

Which is really only relevant if it's ever found. The ocean is HUGE.

melling2• 12 years ago

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.

JacobAldridge2• 12 years ago

"Equivalent to finding a pen in San Francisco" - http://www.cockeyed.com/science/flight_370/flight_370.html

reitzensteinm2• 12 years ago

Ever is longer than the ocean is huge :)

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.

cnvogel3• 12 years ago

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,"""

gsnedders• 12 years ago

{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.}

They're read frequently during maintenance. They are required to be checked to be operable.

kyrra• 12 years ago

Link does not work on mobile (at least for my iPhone).

bobbles• 12 years ago

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."

chris_va• 12 years ago

Random question: Why do black box pings stop after 30 days of 1 ping/second, instead of an exponential tail off for the next few decades?

damian2000• 12 years ago

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.

cnvogel3• 12 years ago

This video has a teardown of an old flight data recorder (tape based), with a ultrasound pinger/locator beacon(at about 3 minutes into the video).

http://youtu.be/mQehX0rVYuY?t=3m48s

Here's some technical information about this particular beacon.

http://www.sea-avionics.com/lc/cart.php?target=productDetail...

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...)

nikatwork• 12 years ago

Assuming the conclusion is correct, what would be the most likely explanation? Pilot murder/suicide?

m-app• 12 years ago

Might be a lack of oxygen, similar to Helios Airways 522?

http://en.wikipedia.org/wiki/Helios_Airways_Flight_522

leoedin• 12 years ago

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.

Zigurd• 12 years ago

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.

mfsampson• 12 years ago

Ah, nice. A factual document about this. Can't believe they have already brought out books about this when the aircraft is still missing.

parados• 12 years ago

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.

brc• 12 years ago

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?

ximeng• 12 years ago

Malaysia and Australia sharing about 50 million dollars.

hga• 12 years ago

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).

phkamp• 12 years ago

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.

wglb• 12 years ago

Do you have any sense of an alternative to their very shaky assumption, or are they essentially boxed in by the available numbers?

goshx• 12 years ago

TLDR: we still don't know where it is.

The Search for MH 370 (Inmarsat)

Related Stories