Cooling Idea

[The Doctor77]

So made an account just to ask this question, but idk if this idea has been thought of before and my apologies if it has. After watching the new desk pc videos, i kinda had an idea for a cooling solution for computers, the idea of a desk pc has always been kind cool to me, but what if you made it air tight, pulled a vacuum on it and got rid of all the moisture in the system, and filled it with a single type of gas, like nitrogen or hydrogen, something without any moisture in it, then ran it through some sort of chiller, would you be able to get sub ambient temps without condensation on the parts? is it even possible? has it been done before? if so how'd it go? just thought i'd try starting a topic about this, if its not been done before that is.

[char3327]

It sound possible, but the application of the idea sounds impractical... cool idea though! That would be sick if you could get it to work

[Asc1005]

18 minutes ago, The Doctor77 said:

So made an account just to ask this question, but idk if this idea has been thought of before and my apologies if it has. After watching the new desk pc videos, i kinda had an idea for a cooling solution for computers, the idea of a desk pc has always been kind cool to me, but what if you made it air tight, pulled a vacuum on it and got rid of all the moisture in the system, and filled it with a single type of gas, like nitrogen or hydrogen, something without any moisture in it, then ran it through some sort of chiller, would you be able to get sub ambient temps without condensation on the parts? is it even possible? has it been done before? if so how'd it go? just thought i'd try starting a topic about this, if its not been done before that is.

but the thing is different gases have their own downsides for example hydrogen is a very flammable gas and if somehow anything shorts and creates a spark inside then the desk will go boom.

 

and other thing is pressure to maintain because diff gases react different to temperature and some small pressure differences can lead to accidents. 

[Freakwise]

Well it's possible and even "achievable" for average user to certain degree (if they Willing to do it), just not with a desk that has glass top

 

Or any desk for that matter...

Edited September 15, 2021 by Freakwise

[Guest]

45 minutes ago, The Doctor77 said:

So made an account just to ask this question, but idk if this idea has been thought of before and my apologies if it has. After watching the new desk pc videos, i kinda had an idea for a cooling solution for computers, the idea of a desk pc has always been kind cool to me, but what if you made it air tight, pulled a vacuum on it and got rid of all the moisture in the system, and filled it with a single type of gas, like nitrogen or hydrogen, something without any moisture in it, then ran it through some sort of chiller, would you be able to get sub ambient temps without condensation on the parts? is it even possible? has it been done before? if so how'd it go? just thought i'd try starting a topic about this, if its not been done before that is.

You cant really vacuum seal a PC due to the I/O ports not being sealed, (for example gas could escape through the USB ports.)

[LWM723]

Use hydrogen and post the results if you're still around after pushing the power switch.

[suedseefrucht]

2 hours ago, The Doctor77 said:

what if you made it air tight, pulled a vacuum on it and got rid of all the moisture in the system, and filled it with a single type of gas, like nitrogen or hydrogen, something without any moisture in it, then ran it through some sort of chiller

Let's start with a fact:

Water conducts heat better than nitrogen or dry air.
 

So if you remove the moisture, the heat transport will be worse than before.

In your closed system, the hot air can't go anywhere, so you need the chiller.

Because air conducts heat much worse than water, you will need a very big chiller or you need to move the air through the chiller very slowly.

So it will be like water cooling but worse, because you use gas instead of water and gas is a worse heat conductor.

 

So as constructive feedback, just do watercooling or let some fans exchange the air inside your pc.

 

2 hours ago, The Doctor77 said:

would you be able to get sub ambient temps without condensation on the parts?

That's the only reason to pick your solution over watercooling. But because of the bad heat conducting of gas, the chiller would need to be big and you would need something like a heat pump with a compressor. The compressor will need extra energy and the extra energy will produce extra heat in your room.

So you may get below ambient, but with more heat in your room, more power consumption, more noise from the compressor and a big box next to your desk.

 

Good air cooling or watercooling is enough to avoid thermal throttling.

So you will have to decide if it's worth the efford just for a number in your temperature log.

 

In my opinion, a disadvantage of the usual desk pc is the very thick desk top plate. You will either hit your knees against it, or your elbows will be too high. So I recomend to make it less than 60 mm thick. This is my solution for a pc desk with a thin desk top plate:
PC desk opinions - Cases and Mods - Linus Tech Tips

[The Doctor77]

19 hours ago, char3327 said:

It sound possible, but the application of the idea sounds impractical... cool idea though! That would be sick if you could get it to work

thanks, tho as suedseefrucht mentioned the idea kinda falls through, it was never really something i was considering to do, just sorta an idea that if possible i might attempt sometime in the future when i had the spare cash lying around. definitely impractical though, just a kinda flex thing really.

[The Doctor77]

19 hours ago, Asc1005 said:

but the thing is different gases have their own downsides for example hydrogen is a very flammable gas and if somehow anything shorts and creates a spark inside then the desk will go boom.

 

and other thing is pressure to maintain because diff gases react different to temperature and some small pressure differences can lead to accidents. 

yeah so hydrogen would be out of the question, and to do it right then there would need to be a some pressure regulators with maybe some kind of reservoir but for gas. 

[The Doctor77]

19 hours ago, Freakwise said:

Well it's possible and even "achievable" for average user to certain degree (if they Willing to do it), just not with a desk that has glass top

 

Or any desk for that matter...

just curious but why might that be the case? 

[The Doctor77]

19 hours ago, Cool_Evlo said:

You cant really vacuum seal a PC due to the I/O ports not being sealed, (for example gas could escape through the USB ports.)

true that is a problem, but that would have to come under one of those things where you'd prep the system by personally sealing the ports, or you could opt for wireless and just pre install the dongles before sealing it up. god help you if something isn't done right the first time though.

[The Doctor77]

19 hours ago, LWM723 said:

Use hydrogen and post the results if you're still around after pushing the power switch.

ahaha yeah, Asc1005 mentioned the whole explosiveness of hydrogen, so that would be out of the question, in case literally anything arced or sparked.

[The Doctor77]

17 hours ago, suedseefrucht said:

Let's start with a fact:

Water conducts heat better than nitrogen or dry air.
 

So if you remove the moisture, the heat transport will be worse than before.

In your closed system, the hot air can't go anywhere, so you need the chiller.

Because air conducts heat much worse than water, you will need a very big chiller or you need to move the air through the chiller very slowly.

So it will be like water cooling but worse, because you use gas instead of water and gas is a worse heat conductor.

okay first off, i did a quick search and found this article:https://www.electronics-cooling.com/2003/11/the-thermal-conductivity-of-moist-air/, it has a graph in there that actually contradicts what you're saying, now i admit im kind of a dumbass so correct me please, i might be reading that wrong, but if it is true, dry air has a higher w/mK then humid air, and humid air even drops off in conductivity past a certain temperature. and with the chiller i'd have the internal air run over a radiator thats cooled externally, sorta like a cars ac system, if that makes sense. 

 

17 hours ago, suedseefrucht said:

So as constructive feedback, just do watercooling or let some fans exchange the air inside your pc.

this idea started off as just a niche little chat thing, i never actually intended to use this option over conventional aio cooling, and if i ever were to attempt it, it would be a proof on concept rather than an actual drop in replacement to my computer now, i probably should have specified that in the original post sorry.

 

17 hours ago, suedseefrucht said:

That's the only reason to pick your solution over watercooling. But because of the bad heat conducting of gas, the chiller would need to be big and you would need something like a heat pump with a compressor. The compressor will need extra energy and the extra energy will produce extra heat in your room.

So you may get below ambient, but with more heat in your room, more power consumption, more noise from the compressor and a big box next to your desk.

that was the whole reason for pitching this idea in the first place, the sub-ambient temp possibility, also, if you're going to the extent of making a custom atmosphere for your pc, chances are you wont mind sticking the chiller and the noisy/hot bits outside and just pump the stuff through the wall or a window. sorta like linus' whole room water cooling thing that he did. again this was only ever meant to be a niche idea, not a viable commercial drop in replacement to current cooling solutions today, just an overclockers peace of mind kinda thing i supose.  

 

 

 

[Freakwise]

50 minutes ago, The Doctor77 said:

just curious but why might that be the case? 

To many spot for leakage point with glass and wooden desk

[The Doctor77]

2 minutes ago, Freakwise said:

To many spot for leakage point with glass and wooden desk

ahh righto, what if the desk, instead of being made with wood, made it with metal, and glass is saleable, with some kind of silicon, or make the whole thing out of metal and then do what this guy did and lay veneer down to give it a wooden look to it?

[Freakwise]

2 minutes ago, The Doctor77 said:

ahh righto, what if the desk, instead of being made with wood, made it with metal, and glass is saleable, with some kind of silicon, or make the whole thing out of metal and then do what this guy did and lay veneer down to give it a wooden look to it?

That could work technically speaking, but there's some overall general problem with this, as bad as wood is at thermally conducting, over time (1-2 hrs) the cold Will "crept out" into the surface and surrounding of the desk

 

You can however mitigate this by using some sort of thermal shield material in between wood and metal, but it just something to keep in mind

[The Doctor77]

33 minutes ago, Freakwise said:

That could work technically speaking, but there's some overall general problem with this, as bad as wood is at thermally conducting, over time (1-2 hrs) the cold Will "crept out" into the surface and surrounding of the desk

 

You can however mitigate this by using some sort of thermal shield material in between wood and metal, but it just something to keep in mind

yeah thats true, you'll end up with no condensation in the desk but the top of it will end up getting a bit slippery

 

[suedseefrucht]

5 hours ago, The Doctor77 said:

okay first off, i did a quick search and found this article:https://www.electronics-cooling.com/2003/11/the-thermal-conductivity-of-moist-air/, it has a graph in there that actually contradicts what you're saying

Wow, that's surprising. Water is a better thermal conductor than air but water vapor is worse. Good to know!