Immersion cooling

[Kevo05s]

Here you go guys,

 

I thought this was interesting concept...

Youtube video of how it works

Also, this got me thinking, we might be able to use this in a pumpless loop...

bare with me: if all the blocks had their components on the side,and had the input at the bottom and output at the top, and we would run them in parallel, the liquid could arrive at the bottom of the block, then go upward to the radiator at the top, with a slight angle (the vapor pipe would come in at the highest point, and the lowest point would be where the condensed liquid would go out, and there would be a reservoir halfway in the middle of the going back down process

I know I'm simple minded on this one since there's a lot of info missing here about the physics properties of this liquid but its more of a concept, an idea...

 

The attached picture is what it would look like (using the red lines as pipes) 

 

[RadiatingLight]

Water doesn't go up...

[vorticalbox]

You would need a way to cool the radiator below ambient in order to condense the liquid back well into a liquid. 

[Mira Yurizaki]

You're going to need to run the entire system at a really low pressure. I mean, I'm sure I don't have to tell you that not only does water not become a vapor until 100C at standard atmospheric pressure, but it needs a lot of heat to still turn it into vapor. i.e., water at 100C doesn't instantly become vapor. If you don't want to do that, you could use a liquid with a lower boiling point like alcohol.

 

The only problem is you have to make sure the system is tightly sealed. Just because it's liquid tight doesn't mean it's vapor tight.

[Kevo05s]

You guys missed the point here, this is not water, it's 3M Novec, a lab made liquid that boils at just above room temp

[Kevo05s]

36 minutes ago, RadiatingLight said:

Water doesn't go up...

first, this wouldnt use water, it would use 3M Novec, and it evaporates, while vapor does go up

[sl06bhytmar]

Then you should first make sure your loop is AIR TIGHT, so you need to get a hand pump and connect it to your loop system with pressure gauge to see how well it keeps pressure. If you see any pressure drop within 24 hours then it is not feasible with your current loop (air is leaking).

 

As stated in above, water tight doesn't mean stuff is air tight. Air tight means water tight because water molecules are larger than air molecules.

[0ld_Chicken]

I think this could work with custom blocks and some other things that I'm sure I'm not thinking of. 

 

also not sure if you've seen this video of Novec.  The vapor is cooled by the radiator at the top which is just a regular loop

 

[Kevo05s]

42 minutes ago, 0ld_Chicken said:

I think this could work with custom blocks and some other things that I'm sure I'm not thinking of. 

 

also not sure if you've seen this video of Novec.  The vapor is cooled by the radiator at the top which is just a regular loop

 

yup, seen that, its a part of the reason why I thought it could work for consumer grade

 

48 minutes ago, sl06bhytmar said:

Then you should first make sure your loop is AIR TIGHT, so you need to get a hand pump and connect it to your loop system with pressure gauge to see how well it keeps pressure. If you see any pressure drop within 24 hours then it is not feasible with your current loop (air is leaking).

 

As stated in above, water tight doesn't mean stuff is air tight. Air tight means water tight because water molecules are larger than air molecules.

yeah I understand, and this "loop" I showed isnt going to be built anytime soon, there's some factors we dont know like how would the Novec react to O-rings... mineral oil doesn't like rubber, who says this will like it? Plus my idea would work better if it was manifactured by a professional, not Mr nobody like me

[Stefan1024]

On 2016-11-17 at 9:09 PM, Kevo05s said:

Here you go guys,

 

I thought this was interesting concept...

Youtube video of how it works

Also, this got me thinking, we might be able to use this in a pumpless loop...

bare with me: if all the blocks had their components on the side,and had the input at the bottom and output at the top, and we would run them in parallel, the liquid could arrive at the bottom of the block, then go upward to the radiator at the top, with a slight angle (the vapor pipe would come in at the highest point, and the lowest point would be where the condensed liquid would go out, and there would be a reservoir halfway in the middle of the going back down process

I know I'm simple minded on this one since there's a lot of info missing here about the physics properties of this liquid but its more of a concept, an idea...

 

The attached picture is what it would look like (using the red lines as pipes) 

 

Keep the high exansion from liquid to vapor in minde. Otherwise you get quickly a very high preassure in the system and the sealings will break. Also a normal radiato has very narrow channels, not sure it this works with vapor.

[PakCikKean]

I was having the exact same thought when I found out what Novec was. On paper it sounds incredibly useful and efficient since it doesn't require a pump. All it needs (I think, don't quote me on this), a radiator and an incredibly air tight case. The problem is.... it needs to be air tight. When you have cables and wires running out from your case, it means that you need a hole (or multiple) to run your cables through and that would mean chances for leaks to happen. Since Novec evaporates so easily, having leaks means you're releasing it to the atmosphere hence have to refill your tank quite often. Until someone designs a case that could negate this problem, Novec could be the future of PC cooling.