Explaining the no pump liquid cooler.

Akai Kiri · March 22, 2019

The video you guys made about this finally got me to sign up to the forum. I have been ~~stalking you~~ lurking for years.

The cooler works on some pretty simple principles. As we all should know, water seeks its own level. Every time Linus doesn't want to drain the loop we can see this in action. Pressure is also at work here. When the fluid turns into a gas the pressure tries to equalize. the radiator is cooling some of it and it turns back into a liquid so there is always a difference in pressure. In other words, the boiling point of the fluid is low enough that it is able to carry heat away from the CPU and dump it into the radiator before flowing back into the loop as a liquid. water blocks will need to be designed differently for computers that stand up or lay down to make use of this. You don't want the coolant input to be messing with the vapor output which means the water level needs to be just right. (strategic reservoir positioning intensifies)

The fluid could be a number of things but the idea here is that it boils at a low enough temp to be converted to a gas by the cpu and also be cooled enough to be a liquid again. if the boiling point is too high it will not make it into the radiator and then you don't get circulation and thus no cooling. you also don't want it to be too low because the cpu under load could turn all of it or most of it into a gas and its the same result.

If i had to guess it would say the fluid inside is actually acetone or Bromine. Acetone boils at 50.5C and Bromine at 58.8c. Chloroform boils at 62.2c but unless this is an well designed Chloroform smuggling operation i feel like thats a bad choice. more info here -> https://www.engineeringtoolbox.com/boiling-points-fluids-gases-d_155.html

I have a few suggestions on things you can try to get the best cooling effect possible.

1, try insulating the output from the CPU. It may seem stupid to do so but you want to make sure that as much of the fluid makes it away from the cpu as possible so it doesn't re condense in the tube.

2, try positioning the radiator so that the output from the cpu is slightly higher than the input to the radiator to ensure that gas only flows out of the cpu and into the radiator in one tube. Remember, pressure equalizes and shit flows down hill.

While you cant really change the radiator because everything is sealed i want to suggest to whoever made that thing to use a radiator that has a tube all the way through it and not just one with a small res on one side. the difference in normal water cooling (from what i have read) is not much different between the two designs. In this situation you don't have nearly as much coolant in the loop AND it contacts the cpu longer than a traditional liquid cooling system so you need to cool it as best you can. Unless im simply wrong, i don't think its possible to cool a cpu below the temperature of the coolant so a few degrees lower on the input coolant will have a great impact on temps.

BONUS SECTION

this would be a great project for sketchy heatsinks.