Google NexusOne Can Survive a Hard Vacuum | Hacker News Reader

Google NexusOne Can Survive a Hard Vacuum

technologyreview.com

60 pointswglb21 comments

The vacuum isn't what I'd worry about; I'd worry about the radiation.

I thought that space stuff usually used ancient CPUs because they had to be rad-hard in ways that stuff beneath the ionosphere doesn't need to be, and creating a sufficently rad-hard CPU was difficult and time consuming (thus you end up with what was state of the art when you started working on it, some time ago)

True but there's also a difference, you're not controlling a multi-billion dollar piece of hardware, perhaps containing astronauts; your tolerance for failure due to radiation can probably be higher for a hobbiest science experiment.

Plus, using an off-the-shelf device means there's a whole bunch of stuff you don't have to build from scratch - even if it also contains a whole bunch of stuff you wouldn't have built if you did it from scratch.

agreed. but the radiation in low-earth orbit, at least at the altitude they seemed to have shot the phones, is usually not intense enough to destroy commercial electronic devices (instantly).

we just finished a radiation susceptibility test for a commercial IC supposed to fly on a satellite in low-earth orbit (which you can find in a lot of consumer electronic products): no serious performance degeneration occured while exposed to a much higher total dose than expected during the lifetime of the satellite.

EDIT: if you're interested in the effects of radiation on electronic devices/components in low-earth orbit, check [0].

[0] http://klabs.org/DEI/References/design_guidelines/design_ser... [pdf]

Yeah, I'm sure almost any phone will survive a hard vacuum.

Why are you sure? Things that could go wrong I can think of:

    - capacitors leaking.
    - gas bubbles in various parts.
    - heat problems due to lack of convective cooling.

can you not put shielding AROUND the device, rather than build hardened processors ?

I can think of one reason - increases weight of the whole package - but perhaps the corresponding cost increase is not so much as buying a radiation hardened 1Ghz processor.

One of the interesting things is that there is some really high energy particles running around in space. If you have a lot of shielding, you risk these particles hitting that shielding causing a cascading effect making things worse (My father designed satellites and we used to chat about this stuff when I was a kid).

you can't shield your devices from every type of radiation up there with a reasonable sized shield. so yes, it's about weight.

Agreed: One sunspot or gamma ray burst and your Nexus One will stop working very rapidly.

The "age" of rad hardened devices is pretty much limited by the processes used. Only simpler designs work well. The finer the traces, the more charge-related problems occur and the less reliable they will be. Also, there's not much interest in making it better because it works fine as it is.

If you take away the JVM, most of the OS, the graphics stack, comms layer and all the fancy abstractions from the Nexus One and Android OS, a 1MHz RCA 1802 is fine ;-)

Simple is good enough though when you read a bit of telemetry and you use low bandwidth RF comms.

Also the Nexus is a really bad idea as it contains FLASH, which if you fart any radiation near, it flips bits like crazy.

Cool article about nanosatellites , but the HN title needs to be more descriptive.

Nitpick: LEO is not a hard vacuum, it's a medium vacuum - it's not even considered a high vacuum.

LEO has enough air that satellites experience noticeable drag and friction from it.

The article mentions GPS, but doesn't US law require the GPS to disable itself once reaching a certain speed and altitude?

I wonder how they get around that.

No, US law requires no such thing.

Receivers capable of operating above 18km and 515 meters/sec are export controlled, but then, so is cryptography software.

Apparently they are not allowed to operate at altitudes higher than 60000 ft and speeds greater than 1000 knots.

http://blog.jgc.org/2010/11/gaga-1-cocom-limit-for-gps.html

That blog entry fundamentally misunderstands the restriction.

The relevant document is available here: http://www.armscontrol.org/documents/mtcr

Note that it refers to export controls, not any blanket ban on functionality.

Given that phones are sold worldwide and phone vendors probably don't want to deal with ITAR paperwork for features that aren't even officially supported, I'm betting they have crippled GPS.

HTC is a Taiwanese company. Their R&D and manufacturing are all done in Asia. They import phones to the US, they don't export them from the US. ITAR has rather limited relevance to them.

There's a decent chance (although it is not necessarily the case) the GPS receiver chips have some US-originated IP in them.

That would be irrelevant for export regulations though. A better reason for staying inside the export regulations, is that it would make it problematic to reexport it, so e.g. amazon would have to apply for export permit to sell it from the US and other countries that have signed the Wassenaar agreement, which is most western countries.

Sounds like they could have used a blackbox.

Reply on news.ycombinator.com