Stories of computers in the antarctic, in space etc.

[tashirosgt]

Another 100 F plus degree day here in southern NM. This is typical weather, but it would be nice to hear some stories about using computers in conditions of extreme cold. Do computers in devices on the exterior of the International Space Station have to have heaters in them? Perhaps someone can tell about using a hair dryer to unstick power supply fans or using blowtorches to warm up motherboards - stuff like that.

[Evil_Bert]

Not sure about what goes on in space, but electronics modules I've used for "extreme conditions" usually have sealed enclosures, anti-icing heaters inside, humidity control valves, desiccant packs, etc. A couple I've worked with had permanent pressurisation from an "environmental unit" that supplied pressurised, dried air at around 15-20 deg C.

[Dan]

A bit off topic ... but ... what part of Southern NM?

[tashirosgt]

Las Cruces, where people from Arizona come for the cooler weather.

[Dan]

Ah! I know the place. Home of UNM, the UAV testing center, RGM Communications, Mesilla Valley Produce, Zianet, and also an active participant in the annual Southwestern Soil Exchange Program, whence once a year -- from about the end of February to the end of June -- mother nature collects all the fertile topsoil in southern NM, sends it into Texas, and imports a fresh batch of desert dirt from Arizona!

[tashirosgt]

True about the soil exchange. But it's NMSU. UNM is further north. Zianet was taken over by La Tierra Communications but the zianet.com domain lives on. RGM Communications, I don't know about.

[DBelton]

I do know that electronics that are exposed to extreme temperatures have to have some type of insulation/humidity control or else they don't last very long.

Just for an example, years ago, I worked for a material handling company servicing electric lifts. There was one company that made ice cream, and had a lift that was constantly in and out of their freezer that was around -20 F. We tried to sell them the control module that is sealed for extreme environments, but they didn't want to spend the extra money at the time.

Well, less than 1 month later, we were dragging the lift out of the freezer because all of the electronics were iced up and frozen solid.

Took 2 cases of Blue Shower (at over $20 per can) to get that lift thawed out and working again.

After we hit them with the bill for that, they wished they had spent the little extra for the sealed control module.

[stevea]

I've been through the space electronics design issues before, tho' the stuff I worked on more recently was for a 'high altitude' (20km) typical working altitude.

In space if your electronics are in the sun you have a severe problem with over-heating. There are ways to manage heat by dissipating it to the dark side, but still - without an atmosphere to convect heat you are stuck with conduction from the power-users and ultimately you have to consider radiation as the main means of sourcing/sinking heat.

If you have a perfectly conductive black sphere in the same orbit as the earth it reaches a cool equilibrium temo ~7C or ~45F IIRC. But without the Sun it approached the ambient 4 Kelvin space ambient temp (-269C roughly -465F). In the Sun on a less than perfect conducting ship, it's not hard to get temps of 200C or ~400F temps in practice. So reflective white is the new black at least for your electronics bays. Yes they may require heaters, but often the electronic is on 24x7 and provides it's own heat.

In the high altitude case the temp swings are only a little less brutal, but even at 20km you have a little air (about 1/15th surface pressure). Something over 100C on the top/hot side in the sun and -100C on the top(dark) side at night. The bottom side toward the earth does dramatically better. At night the ambient earth radiation (earth shine?) means the bottom side is much warmer. (don't recall a figure). You might guess the low pressure means convection is ignorable, but that's not true. The heat capacity of the low-pressure atmosphere is less, but so long as the mean-free-path of the molecules of the atmosphere are small then you are heating the mass of the gas and not just the rare impinging molecules. So even at 20km alt blowing the thin air over your heat sinks makes a lot of sense. You have to get over 100km, into low earth orbit range) before the air convection becomes ignorable.


Yes it's cold in space - except when it's hot.

---

The radiation in space is another issue - and the main reason these multi-year missions to Mars are currently a pipe-dream or else a suicide mission. All the space stations are protected by the Van Allen belt. Only the week-longish lunar excursions have we had humans outside that protection. Manned Mars mission idea idea "needs work".

[Dan]

True about the soil exchange. But it's NMSU. UNM is further north. Zianet was taken over by La Tierra Communications but the zianet.com domain lives on. RGM Communications, I don't know about.

Yikes! My bad. You're right about NMSU. I was typing faster than I was thinking. Although the War Department got quite the chuckle out of it. (They signed my paychecks for several years.) <....>

[beaker_]

Best little computer I've bought. It has seen both +40C and -40C, survived >6atm, blown off the roof of a truck box at 100kmph, dropped from 30', stepped on countless times, and has never stopped ticking.

[stevea]

Originally Posted by beaker_

Best little computer I've bought. It has seen both +40C and -40C, survived >6atm, blown off the roof of a truck box at 100kmph, dropped from 30', stepped on countless times, and has never stopped ticking.

And is doesn't run Gnome3 !

[beaker_]

Originally Posted by stevea

And is doesn't run Gnome3 !

Thank God No..!

[tashirosgt]

I'd think that electronic components would contain some trapped air and moisture. Why don't they burst in a vacuum? Why doesn't freezing and thawing crack them like it cracks concrete? - or perhaps it would.

[flyingfsck]

In North America, you don't have to go far to get trouble with heat and cold. Laptop computer screens typically stop responding at about -20 Celsius (but it will recover when brought back inside), while the heat of Colorado can kill it.