overclocking Overclocking in Vacuum chamber

[Arshia]

Have anyone done it before? If not (I haven’t heard anybody doing this) why? As I know the biggest concern in subzero over locking is condensation. Wouldn’t sucking air from components solve it? If

 cooling less heat generating parts is necessary can it be filled up with some gasses that won’t cause issues? Are there any other problems like bending duo to different materials contraction holding people back from -196C (-195.8) over clocking?

[AbydosOne]

If you don't have any air, there's nowhere for heat to go (radiation is much slower).

[Arshia]

1 hour ago, AbydosOne said:

If you don't have any air, there's nowhere for heat to go (radiation is much slower).

As I said you can fill it with other gasses that don’t conduct electricity in liquid from

[Master Disaster]

1 minute ago, Arshia said:

As I said you can fill it with other gasses that don’t conduct electricity in liquid from

That's an oxymoron, if you replace the air with other gasses then it wouldn't be a vacuum.

 

Either way it doesn't matter anyway, if you take something very hot and place it inside somewhere very cold it condenses the gas around it with causes condensation. You'd need something that remains as a gas at the hottest temp your device can reach to not cause condensation and at that point you're not getting any cooling from it anyway.

 

You place a computer in a vacuum and it fries, air (or gas) is what transfers the heat from the device to the environment, no gas means no heat transfer. You ever seen how you can dip your hand in water the very quickly plunge into molten metal and suffer no injury? The reason is because the water very quickly turns into steam which forms an insulating layer around your hand protecting it from the heat, think of that effect but in reverse.

[ShrimpBrime]

2 hours ago, Arshia said:

Have anyone done it before? If not (I haven’t heard anybody doing this) why? As I know the biggest concern in subzero over locking is condensation. Wouldn’t sucking air from components solve it? If

 cooling less heat generating parts is necessary can it be filled up with some gasses that won’t cause issues? Are there any other problems like bending duo to different materials contraction holding people back from -196C (-195.8) over clocking?

Condensation really depends on the dewpoint. You can have very little or a lot of condensation.

 

It's not practical to overclock your PC in a vacuum chamber either. If you want to swap out memory in the middle of a run, then you have a bunch of work ahead of you. 

 

LN2 would work in a vacuum chamber because it doesn't rely on atmosphere to be cold. In fact nitrogen when boiling off creates it's own atmosphere and can be dangerous to people in a small non ventilated room. 

 

I might be mistaken, but as LN2 transforms from liquid to gas, you'd have to keep the vacuum pump running to keep the vacuum in the chamber.

 

Pretty much sealing the board from condensation is most practical and normal proceedure, but LN2 in a vacuum is curious and probably possible to do.  

[Spindel]

Theoretcally you could put the MB with CPU in a vaccum chamber. Then run hosing for a sub ambient brine through the wall of the vaccum chamber. Only condensation you risk is on the hosing outside the vaccum chamber, but non where you risk any electronics. 

[wkdpaul]

vacuum doesn't mean sub zero temperatures.

 

If you're talking about LN2 overclocking in a vacuum, then it's going to be VERY difficult since LN2 evaporates, and so the vacuum chamber wouldn't be in a vacuum anymore.

 

Dealing with condensation when sub-zero cooling is easier than trying to rig a vacuum chamber that would hold vacuum with LN2 in it.

[Cyracus]

Love y'all that are concerned about the ln2 off-gassing in the vacuum and not how to add more ln2

[wkdpaul]

28 minutes ago, Cyracus said:

Love y'all that are concerned about the ln2 off-gassing in the vacuum and not how to add more ln2

The LN2 off gassing is only one of the hundreds of issues this idea has. It's a bad idea all around, and the OP seems to have very little understanding of how vacuum, heat transfer, and many other things work (he mentioned vacuum, but also that "air" should be replaced with gas that wouldn't allow for condensation ... r i g t h ...).

[Arshia]

2 hours ago, wkdpaul said:

vacuum doesn't mean sub zero temperatures.

 

If you're talking about LN2 overclocking in a vacuum, then it's going to be VERY difficult since LN2 evaporates, and so the vacuum chamber wouldn't be in a vacuum anymore.

 

Dealing with condensation when sub-zero cooling is easier than trying to rig a vacuum chamber that would hold vacuum with LN2 in it.

I didn’t say vacuum means subzero temperature. I was thinking of a custom lid that attaches to cpu on one side and has LN2 pod on the other side which is in open air it’s outside of chamber. 

[akio123008]

6 minutes ago, Arshia said:

I didn’t say vacuum means subzero temperature. I was thinking of a custom lid that attaches to cpu on one side and has LN2 pod on the other side which is in open air it’s outside of chamber. 

Why don't you just build it and post a picture of it here

[Arshia]

4 hours ago, Master Disaster said:

That's an oxymoron, if you replace the air with other gasses then it wouldn't be a vacuum.

 

Either way it doesn't matter anyway, if you take something very hot and place it inside somewhere very cold it condenses the gas around it with causes condensation. You'd need something that remains as a gas at the hottest temp your device can reach to not cause condensation and at that point you're not getting any cooling from it anyway.

 

You place a computer in a vacuum and it fries, air (or gas) is what transfers the heat from the device to the environment, no gas means no heat transfer. You ever seen how you can dip your hand in water the very quickly plunge into molten metal and suffer no injury? The reason is because the water very quickly turns into steam which forms an insulating layer around your hand protecting it from the heat, think of that effect but in reverse.

Yeah it wouldn’t be vacuum anymore but if it’s filled with a gas that doesn’t  conduct electricity when it’s liquid condensation won’t kill any components.  Just sucking air out of a chamber with a custom metal lid that is attached to cpu/gpu on one side and has an integrated LN2 pod and filling the chamber with that gas shouldn’t be that hard. Right?

[Arshia]

4 minutes ago, akio123008 said:

Why don't you just build it and post a picture of it here

I don’t have equipments or time to make it

[akio123008]

5 minutes ago, Arshia said:

if it’s filled with a gas that doesn’t  conduct electricity when it’s liquid condensatio

I think you may not have a correct understanding of condensation; it's not the air itself that turns liquid, it's the water vapor in the air that turns into liquid water.

 

Other gasses may contain water vapor as well and have the same issue. Even if they don't, it's still less effort to just remove water vapor from normal air, than it is to fill a chamber with some sort of other gas.

[wkdpaul]

2 minutes ago, akio123008 said:

I think you may not have a correct understanding of condensation; it's not the air itself that turns liquid, it's the water vapor in the air that turns into liquid water.

 

Other gasses may contain water vapor as well and have the same issue. Even if they don't, it's still less effort to just remove water vapor from normal air, than it is to fill a chamber with some sort of other gas.

Yeah, I think what the OP is looking for is an air conditioning unit.

[Letgomyleghoe]

3 hours ago, ShrimpBrime said:

Pretty much sealing the board from condensation is most practical and normal proceedure, but LN2 in a vacuum is curious and probably possible to do.  

you would need some way of feeding it in then along with being able to torch the pot though. I guess you might not need to torch the pot, but i could see the benefit of using a vacuum chamber in the first place be replaced by the hurdles you would have to go through. 

[akio123008]

10 minutes ago, wkdpaul said:

Yeah, I think what the OP is looking for is an air conditioning unit.

"so is that air conditioner on your desk cooling the computer?"

"no, that's just for dehumidification, hold on a second, let me just get my tools I need to refill the LN2"

[Vitamanic]

A vacuum isn’t a good conductor of heat. Which is a good thing, because if it was, life on earth would be eradicated.

[the gamer that is bad]

7 hours ago, Arshia said:

Have anyone done it before? If not (I haven’t heard anybody doing this) why? As I know the biggest concern in subzero over locking is condensation. Wouldn’t sucking air from components solve it? If

 cooling less heat generating parts is necessary can it be filled up with some gasses that won’t cause issues? Are there any other problems like bending duo to different materials contraction holding people back from -196C (-195.8) over clocking?

 

7 hours ago, AbydosOne said:

If you don't have any air, there's nowhere for heat to go (radiation is much slower).

pull a vacuum in the case with external radiators

[the gamer that is bad]

water cooling

 

[Master Disaster]

34 minutes ago, the gamer that is bad said:

 

pull a vacuum in the case with external radiators

Why? This entire concept is flawed, if you have to have an external heat exchanger then the vacuum is totally redundant.

[the gamer that is bad]

to keep condensation out with the chiller unit cooled outside the case

[bowrilla]

If you have just a block around your CPU, you will have 2 issues: 1st you will bend the mainboard because of the pressure gradient 2nd there will be a heat transfer going on and you'd might end up with subzero temps on the PCB around the chamber resulting in condensation.

 

If you put everything inside the box you will have issues with the connections (video cable, keyboard, mouse, etc).

 

The next topic would be: how do you cool to subzero? You can use LN2 or anything similar but that means you will need to constantly fill up the pot. Or you run a loop with a sub-zero coolant. Now you have material issues. You'd need to find a pump, that survives these temps, tubing that doesn't get brittle at these temps, etc. You will get condensation on those though.

 

As mentioned before: it's not the air that's condensating, it's the water in it. Lower the humidity level and you'd get rid of condensation.

 

It's still absolutely pointless though. You can't run such a rig 24/7 without a plethora of issues that need fixing and a ton of money. And for the occassional overclocking competition just insulate your baord with vaseline.

[derr12]

you'd be better off leaving the sealed box at 1 bar, but removing ALL humidity from it somehow.  no moisture in the air = no condensation. pipe in a dehumidifyer and set to maximum powah!