How are computers cooled in space?

[straight_stewie]

As an aside to all of this, computer components on spacecraft don't only need to be cooled, they also sometimes need to be heated. Components on spacecraft need active temperature control. Many components have a relatively narrow temperature band that they can operate in, relative to the extremes of temperatures that can be experienced in space travel and interplanetary exploration anyway.

 

Before I start, I use the term "radiator" to signify a device that uses heat radiation to cool components. Heat radiation does not require a transfer fluid anymore than electromagnetic radiation does. This is opposed to "radiators" as we are used to them here on earth, which are essentially just fluid-to-fluid heat exchangers where the atmosphere is implicitly one of the fluids.

In general, they use simple electric heat elements to heat components, and generally use some kind of fluid loop to take heat away to a "radiator" designed specifically to use radiation to lose heat.

However, this isn't always the case. Well, sort of. While it's almost always the case that there are "radiators" on space craft for cooling (There's only so many ways to make heat "go away" after all), sometimes heat is added by nuclear decay or by piping heat from "always on" components that would naturally generate heat (such as nuclear reactors), as opposed to electric heat generators. 

Additionally, cooling isn't always accomplished by radiators. On many rocket engine designs (especially on later stages of rockets, designed to be used only in space) the engine components are cooled by the fuel (which is usually cryogenic on these stages). A good visual example of this can be seen on the later stages of Falcon 9 rockets, where you can usually visually see the engine bell frost up just before ignition, because they flood it with cryogenic oxygen to pre-cool it before ignition.

 

A third example of cooling is the Lunar Rover's amazing "Wax Phase-Change Thermal Capacitors". The rover would store heat from it's batteries by melting wax. The radiating surfaces for the wax reservoirs would have to be manually uncovered by astronauts after they were done driving, because everything was covered with mylar (a mind blowingly amazing heat insulator) for dust protection.

[Mark Kaine]

On 10/29/2020 at 2:57 PM, ShrimpBrime said:

Copper, read the above posts to yours. Copper is the medium. You dont need air.

I don't know if that has been answered yet but while your logic is sound, it's based on a misunderstanding of how this works I believe. 

 

Yes your cpu gets cooled by the copper, it doesn't need air. 

 

Now the problem is your copper isn't going to get cooled by anything, especially not by the space surrounding it because there are *no* particles to do so... 

 

And within seconds... the copper will melt, and then immediately the cpu will melt. That is how it works. Ie not at all.

 

 

Someone please correct me if I am wrong but I don't think I'm wrong... There simply is no heat transfer in space, not the same way it is on earth at least. 

[Guest]

9 minutes ago, Mark Kaine said:

I don't know if that has been answered yet but while your logic is sound, it's based on a misunderstanding of how this works I believe. 

 

Yes your cpu gets cooled by the copper, it doesn't need air. 

 

Now the problem is your copper isn't going to get cooled by anything, especially not by the space surrounding it because there are *no* particles to do so... 

 

And within seconds... the copper will melt, and then immediately the cpu will melt. That is how it works. Ie not at all.

 

 

Someone please correct me if I am wrong but I don't think I'm wrong... There simply is no heat transfer in space, not the same way it is on earth at least. 

Essentially, from what you're saying is there is nothing is space to dissipate the "heat" from the copper. 

 

I had imagined say a large piece of copper. One end may be warmed by the cpu while the other end would be cool, the "heat" transfer would be within the copper and not outside of the copper. 

(Yes there are other materials that are better than copper for transfer of heat)

 

It would be the same theory I have about running a Cpu passively, just as I'm doing now. 

However I think your point is the heat won't dissipate to the ambient "Air" temp.

While Space has nothing except a temperature, essentially in a vacuum, copper for example can still absorb the temperature, just at a slower rate (If I'm not mistaken).

 

I understand the thought, but if in space "Temperature" wasn't to be any sort of issue, be it hot or cold, the argument for the need to cool or even heat the cpu as mentioned previously wouldn't need to exist. 

 

If there was nothing to create a temperature in space IE: Stars, then the average space temp would be absolute 0 and so would everything else in that space.

 

Now, in a vacuum on earth, a cpu would just keep heating up. However the Ambient temp of 20c doesn't help. The Vacuum ambient temperature is very high compared to an ambient temp of (Deep) space (not near our star for example) where the vacuum temp is near absolute 0 right around 2.5k +/-, which is about -265c. 

 

It's only a speculation on my behalf really. I think to find it very feasible to "cool" a processor in space. Thinking outside the box, not all processors have a 220w TDP of an FX-9590 processor. My proposal would be 50w or less can be doable. And that's not far off a 3600X really. 

 

So can you cool (A / AN) processor (type unkonwn) in space? Well sure, why not?

[Mark Kaine]

1 hour ago, ShrimpBrime said:

Essentially, from what you're saying is there is nothing is space to dissipate the "heat" from the copper. 

 

I had imagined say a large piece of copper. One end may be warmed by the cpu while the other end would be cool, the "heat" transfer would be within the copper and not outside of the copper. 

(Yes there are other materials that are better than copper for transfer of heat)

 

It would be the same theory I have about running a Cpu passively, just as I'm doing now. 

However I think your point is the heat won't dissipate to the ambient "Air" temp.

While Space has nothing except a temperature, essentially in a vacuum, copper for example can still absorb the temperature, just at a slower rate (If I'm not mistaken).

 

I understand the thought, but if in space "Temperature" wasn't to be any sort of issue, be it hot or cold, the argument for the need to cool or even heat the cpu as mentioned previously wouldn't need to exist. 

 

If there was nothing to create a temperature in space IE: Stars, then the average space temp would be absolute 0 and so would everything else in that space.

 

Now, in a vacuum on earth, a cpu would just keep heating up. However the Ambient temp of 20c doesn't help. The Vacuum ambient temperature is very high compared to an ambient temp of (Deep) space (not near our star for example) where the vacuum temp is near absolute 0 right around 2.5k +/-, which is about -265c. 

 

It's only a speculation on my behalf really. I think to find it very feasible to "cool" a processor in space. Thinking outside the box, not all processors have a 220w TDP of an FX-9590 processor. My proposal would be 50w or less can be doable. And that's not far off a 3600X really. 

 

So can you cool (A / AN) processor (type unkonwn) in space? Well sure, why not?

Uhh.. ok, yes, Idk where to start... I'll try. Your logic is sound again but the misconception or misunderstanding is that there's "temperature" in space - but there is not - there are no particles to have any temperature, in other words a "nothing" can't have temperature, right?  So that is the problem your cpu and the copper have particles (lots of it naturally) and you're putting energy into them aka "heat" them up and that energy just has nowhere to go... That means while the materials probably wouldn't just "disappear" the integrity would be destroyed by the heat, and very quickly (it also doesn't help if you used something better than copper, you'd just win a couple of seconds before it all melts) 

Can't explain it any better but the point is that there's nothing to keep any temperatures so it doesn't exist per say, it only happens when you expose something to that space / vacuum.

 

 

Now, I think your idea could work if you'd use for example a *huge* copper or whatever material would be adequate radiator, just like a space station does, I think that should work (through radiation as others have explained). 

 

Now I have no idea how big that radiator would need to be but I also would guess not too big, 1-2m radius maybe, basically just a small satellite basically?

 

(could be wrong about the dimensions, but I think that should work, it's essentially just small "cpu space station / satellite" ) 

[Uttamattamakin]

The short answer is ... HEAT PIPES. 

 

https://www.nasa.gov/mission_pages/station/research/news/heat_pipes.html

 

Quote

In satellites used for communications, global positioning systems, and defense purposes, a heat pipe is the device used to regulate temperature and keep the overall systems operating reliably. A heat pipe is a simple device that can efficiently transfer heat from a hot spot to a cooler remote location without the use of a mechanical pump.

 

[Guest]

21 minutes ago, Mark Kaine said:

Uhh.. ok, yes, Idk where to start... I'll try. Your logic is sound again but the misconception or misunderstanding is that there's "temperature" in space - but there is not - there are no particles to have any temperature, in other words a "nothing" can't have temperature, right?  So that is the problem your cpu and the copper have particles (lots of it naturally) and you're putting energy into them aka "heat" them up and that energy just has nowhere to go... That means while the materials probably wouldn't just "disappear" the integrity would be destroyed by the heat, and very quickly (it also doesn't help if you used something better than copper, you'd just win a couple of seconds before it all melts) 

Can't explain it any better but the point is that there's nothing to keep any temperatures so it doesn't exist per say, it only happens when you expose something to that space / vacuum.

 

 

Now, I think your idea could work if you'd use for example a *huge* copper or whatever material would be adequate radiator, just like a space station does, I think that should work (through radiation as others have explained). 

 

Now I have no idea how big that radiator would need to be but I also would guess not too big, 1-2m radius maybe, basically just a small satellite basically?

 

(could be wrong about the dimensions, but I think that should work, it's essentially just small "cpu space station / satellite" ) 

I guess I have questions then.

 

Do we wait for the cooling device to have a certain temperature?

How cold would the cooling apparatus become after X amount of time in space?

How long would it take a 50w cpu to heat X amount of material at X temp? 

If there's no where for heat to go in space, how does the heat of the sun dissipate to earth?

 

[Mark Kaine]

26 minutes ago, ShrimpBrime said:

If there's no where for heat to go in space, how does the heat of the sun dissipate to earth?

Through radiation. Again there is no temperature per say in space. That doesn't mean radiation can't travel trough it, to the contrary because there's nothing in the way that makes it easier for the radiation to travel through space. The temperature only happens when the radiation hits something. For example a satellite, or planet, etc. 

 

And I can't answer your other questions, I don't know how to calculate this. But basically, like I tried to explain that's how satellites, space crafts etc work they all need some heating and cooling devices to keep them at operational temperatures. 

 

 

Quote

This is why you can insulate your house basically using the air trapped inside your insulation," said Andrew Hong, an engineer and thermal control specialist at NASA's Johnson Space Center. "Air is a poor conductor of heat, and the fibers of home insulation that hold the air still minimize convection."

 

"In space there is no air for conduction or convection," he added. Space is a radiation-dominated environment. Objects heat up by absorbing sunlight and they cool off by emitting infrared energy, a form of radiation which is invisible to the human eye.

 

As a result, insulation for the International Space Station doesn't look like the fluffy mat of pink fibers you often find in Earth homes. The Station's insulation is instead a highly-reflective blanket called Multi-Layer Insulation (or MLI) made of Mylar and dacron.

 

 

Quote

Imagine that "your house was really, really well insulated and you closed it up and shut off the air-conditioning," said Gene Ungar, a thermal fluid analysis specialist at NASA's Johnson Space Center. "Almost every watt of power that came through the electric wires would end up as heat."

This is just what happens on the Space Station. Energy from the solar arrays flows into the ISS to run avionics, electronics ... all of the Station's many systems. They all produce heat, and something has to be done to get rid of the excess.

The basic answer is to install heat exchangers. Designers created the Active Thermal Control System, or ATCS for short, to take the heat out of the spacecraft.

Waste heat is removed in two ways, through cold plates and heat exchangers, both of which are cooled by a circulating water loop. Air and water heat exchangers cool and dehumidify the spacecraft's internal atmosphere. High heat generators are attached to custom-built cold plates. Cold water -- circulated by a 17,000-rpm impeller the size of a quarter -- courses through these heat-exchanging devices to cool the equipment.

"The excess heat is removed by this very efficient liquid heat-exchange system," said Ungar. "Then we send the energy to radiators to reject that heat into space."

 

ISS: 

 

 

Thats all actually quite interesting!

 

 

 

[Deslumo]

Funny thing is that most people never think of this. What if we had a water block, filled it with water, and let the cold temperatures of space freeze the water indefinetly?

[Guest]

2 hours ago, Mark Kaine said:

Through radiation. Again there is no temperature per say in space. That doesn't mean radiation can't travel trough it, to the contrary because there's nothing in the way that makes it easier for the radiation to travel through space. The temperature only happens when the radiation hits something. For example a satellite, or planet, etc. 

 

And I can't answer your other questions, I don't know how to calculate this. But basically, like I tried to explain that's how satellites, space crafts etc work they all need some heating and cooling devices to keep them at operational temperatures. 

 

 

 

ISS: 

 

 

 

Thats all actually quite interesting!

 

 

 

The basis of the thinking was deep space. Not so close to a star where you'd take advantage of the heat. 

 

Of course there's a temp in space. It's about 2.5k out there. 

Of course heat is dissipated by radiation. That's why they call them radiators. lol.

Climate control for humans and cooling a processor are a little different from each other.

 

Your body creates a heat, but yet can be cooled off. The human body we take care of with climate control would freeze solid within a day without an external heat source and special insulation.

 

I don't see your argument being feasible to say you could not cool a processor (of whatever type, ARM maybe?) in outer space.

[Mark Kaine]

1 hour ago, ShrimpBrime said:

I don't see your argument being feasible to say you could not cool a processor (of whatever type, ARM maybe?) in outer space.

No of course you can cool it but you need some kind of device, just an insulated box would do probably, and a radiator. 

 

I was just saying that what you proposed, using just a copper (block I'm guessing) wouldn't work, plus don't forget you need to cool or maybe even heat everything else too, I mean you'd at least need some kind of motherboard and power source, none of that would work without a proper apparatus to protect the parts. 

[Guest]

43 minutes ago, Mark Kaine said:

No of course you can cool it but you need some kind of device, just an insulated box would do probably, and a radiator. 

 

I was just saying that what you proposed, using just a copper (block I'm guessing) wouldn't work, plus don't forget you need to cool or maybe even heat everything else too, I mean you'd at least need some kind of motherboard and power source, none of that would work without a proper apparatus to protect the parts. 

So IHS plate only exposed to space, the rest is contained and powered on. Cpu only.

I suppose the design of the cooler could be a block. Still with fins. Maybe much larger (longer) fins than say a standard heat sink.

 

The heat would still "radiate" into open space. Just at a slower pace. 

Then maybe ponder the wattage converted to BTU an hour. 

Calculate the BTU an hour rate a given material can dissipate. ( In a vacuum, in space ) 

Measure the temperature at the surface of the material and measure how fast it dissipates the ambient temp of space.

 

So you should be able to disspate X amount of heat radiated into space either way. It happens all the time. It's just rocks do it slowly (earth) and the sun does it quickly (Fusion). The heat goes somewhere. (just into space)

 

I wonder what the TDP of the sun is?

 

 

[tikker]

16 hours ago, ShrimpBrime said:

I wonder what the TDP of the sun is?

Practically whatever the temperature of the Sun (or any star for that matter) currently is. Stars are in a way a self-regulating system that balance their gravitational collapse with fusion. If the fusion does not provide enough counter pressure, it contracts and will get hotter. If this outputs enough energy it stabilizes or if it  outputs "too much" (in other words gets "too hot"), the star will expand again to dissipate the energy over a larger area at its surface.

[Orange1]

22 hours ago, ShrimpBrime said:

Of course there's a temp in space. It's about 2.5k out there. 

Out where, the further you are from the sun the colder it is. Also you could be in the shade of the sun behind a planet.

[Guest]

16 minutes ago, Orange1 said:

Out where, the further you are from the sun the colder it is. Also you could be in the shade of the sun behind a planet.

That's exactly what I've been trying to convey the entire time. Thank you.

[Orange1]

1 hour ago, ShrimpBrime said:

That's exactly what I've been trying to convey the entire time. Thank you.

However, there might be a situation where the heat from a planet may affect, like waves of energy or something.

[Guest]

1 hour ago, Orange1 said:

However, there might be a situation where the heat from a planet may affect, like waves of energy or something.

Lots of variables to take into account. 

 

It was a vague question in the first place. 

[Glenwing]

On 10/30/2020 at 10:06 PM, ShrimpBrime said:

Of course there's a temp in space. It's about 2.5k out there.

Temperature is a property of an object. It is based on the average kinetic energy of the molecules that make up the object. The concept of temperature does not apply to vacuum.

 

Objects in space have a temperature, which is based on their distance from the sun, because they lose energy through electromagnetic radiation, and if they radiate energy faster than they receive it from the sun (which is based on distance), their total energy (and temperature) will go down until they are radiating energy equal to the amount they absorb. Therefore, objects in space can get to a low temperature.

 

If you are in thermal contact with one of these cold objects, then you can transfer your heat energy away through conduction. But space itself does not have a "temperature" in the sense that you could stick one end of your heatsink into outer space and get in "thermal contact" with space and get super cooling. Space is not an object, so there would be no heat conduction away from your heatsink.

 

The only way heat escapes in space is through electromagnetic radiation, which is extremely weak compared to conduction. There is a reason vacuum is used in the walls of refrigerators for example; vacuum is one of the best heat insulators available since it eliminates both conduction and convection, leaving radiation as the only form of heat transfer across the gap.

 

When people refer to the "temperature of space" being 2.7 K, they are actually referring to the "temperature" of the cosmic microwave background radiation, which itself is not really a "temperature", but rather a measure of radiation intensity translated into "what temperature an object would settle to if it were in vacuum and being heated only by the CMBR". In other words it's not really the "temperature of space", it's just spoken that way to simplify it for people. That's the temperature that objects would settle to in space if not heated by any stars. But again, it's not like space itself is an object with a temperature of 2.7 K that you can put a heatsink in thermal contact with to get super cooling.

[Guest]

7 hours ago, Glenwing said:

Snip.

Thank you for the well written post!

 

I think you've changed my mind and helped answer the question here!!!

 

So it is now the answer becomes 

 

NO, you cannot cool a cpu in space.

[Sauron]

It's a misconception that heatsinks and radiators need air to function. Heat still radiates away in a vacuum, you just might need a larger surface.

[Guest]

7 minutes ago, Sauron said:

It's a misconception that heatsinks and radiators need air to function. Heat still radiates away in a vacuum, you just might need a larger surface.

I'm back to yes, you can cool a cpu in space now......

 

lol. 

 

 

[BuckGup]

https://www.hpe.com/us/en/newsroom/blog-post/2018/07/spaceborne-computer-still-flying-high-one-year-later.html

 

I suggest reading about spaceborne. Also most people forget that PCs in space need insane radiation shielding compared to earth

[Guest]

Just now, BuckGup said:

https://www.hpe.com/us/en/newsroom/blog-post/2018/07/spaceborne-computer-still-flying-high-one-year-later.html

 

I suggest reading about spaceborne. Also most people forget that PCs in space need insane radiation shielding compared to earth

It's really just a vague yes or no question from the OP.

 

So is it YES or NO? 

 

(I voted yes already..... not that there was a pole or something....)

[Justaphysicsnerd]

Didn't Linus already talk about this in his video regarding Apollo 11's Computers ?

[BuckGup]

2 minutes ago, ShrimpBrime said:

It's really just a vague yes or no question from the OP.

 

So is it YES or NO? 

 

(I voted yes already..... not that there was a pole or something....)

It is YES as we have already done so

[Sauron]

30 minutes ago, ShrimpBrime said:

Thank you for the well written post!

 

I think you've changed my mind and helped answer the question here!!!

 

So it is now the answer becomes 

 

NO, you cannot cool a cpu in space.

 

4 minutes ago, ShrimpBrime said:

I'm back to yes, you can cool a cpu in space now......

 

lol.

 

1 minute ago, ShrimpBrime said:

It's really just a vague yes or no question from the OP.

 

So is it YES or NO? 

 

(I voted yes already..... not that there was a pole or something....)

You don't need to reply to everyone who makes a post in someone else's thread just to say you don't know the answer