Testing Mobile Phones For Controlling Space Missions

An anonymous reader writes "Researchers in the UK are sending an Android handset into space in order to test whether mobile phone chipsets are robust enough to be be used as the basis for controlling future space missions — greatly reducing the cost and weight of spacecraft electronics. 'Once in space, the phone will be bombarded by cosmic and solar radiation, and experience temperatures that veer between extreme heat and cold. A computer on the ground will check whether the phone is able to operate normally in orbit, and if no problems are found the phone will be used to perform tasks usually carried out by the satellite's main avionics computer.'"

Smaller and cheaper electronics...

[xMrFishx]

are smaller and cheaper. That's basically what it says. "We want to use mass produced stuff because it's dirt cheap and made on smaller scales than this expensive rubbish we keep losing by taking a wrong turn at mars. We don't know if it'll work, so we'll send some into space and see what happens, and it will keep that marketeer who keeps asking us what we're really doing busy. He thinks we're working on the iSat. We're just seeing if space-tronics is snake oil or not."

NASA got burned on this, literally

[DCFusor]

Let a bunch of ignorant college kids try and use COTS stuff in their smaller, faster, cheaper, better plans. Almost to a man, they didn't understand certain really important issues, as in -- things that work fine fanless in air burn up in a heartbeat in a good vacuum from their own power when the only way to lose heat is by radiation -- which doesn't remove much till things get very hot indeed. Even micropower opamp chips die in vacuum. After all, there are such things as one watt incandescent bulbs....that get white hot with one watt input, and some of them aren't even in a vacuum! Ok, spread that heat around a 1 watt cpu, which is bigger -- and it's still above the destruction temperature of a chip -- well above.

So, unless they customize the boards for conductive heat removal and some temp control extras, it ain't going to fly. It's been done and evidently the UK guys don't know about it (all too common these days) or don't have a clue what that problem is.

But if they couldn't just buy the parts and make/program their own, they're not smart enough to succeed anyway -- those other problems like bit-glitches caused by radiation and so on will kill them if they don't do a very robust software design with various safeguards and redundancies. Why be stuck with a cel phone circuit board when you could just buy the same parts and add the stuff you really need on the mission all on the same board?

Back in the day, I worked on some stuff that was going into birds. They made us take this class on "What works and doesn't work in space". It was killer enlightening about what the issues are. Some of it has been obviated by new tech -- for example "no electrolytic caps" -- we have ceramics now that serve fine and are probably in most all new tech. "no potentiometers" "absolute minimum connectors" and an entire other course about how things wind up cold welding together in vacuum and most lubes don't work (including surprisingly, graphite which requires an oxygen layer to be slippery). Things like the tempco monster when using dissimilar materials need extra thought so things don't simply warp or explode at big temperature swings as well.

So, NASA has been there, and done that, and even they forgot some of the lessons when they pissed off most of their real engineers and substituted young punk academics with no real world experience...

Here goes history rhyming again.

Re:Smaller and cheaper electronics...

[Locke2005]

Also, if reliability is an issue, a voting cluster of hundreds of small, cheap CPUs may be both cheaper and more reliable than a few expensive mil-spec CPUs... especially since MIL-spec are generally 10 years behind state-of-the-art by the time they are approved.

Re:Smaller and cheaper electronics...

[confused one]

True. But one could (theoretically) send 10's or even 100's of cheap systems into space for the cost of 1 high reliability space rated system. When you need the high-rel system, by all means use it; but, don't let it limit you in such a way as to prevent launching clusters of small cheap satellites or robots when that's an acceptable option. There are bound to be cases where that's of benefit.