Thinking outside the box: watercooling an air cooler

Shawn Knight

Posts: 15,646   +199
Staff member
Through the looking glass: Additional testing and comparisons are needed to determine how this setup would perform against a traditional air cooler, a basic all-in-one watercooling kit or a quality waterblock. Still, it's a neat idea and a prime example of thinking outside the box. And who knows, maybe the concept will catch on?

Deciding between a traditional heatsink and fan combo or a watercooling setup to keep CPU temperatures under control is no easy task. Each method has its own pros and cons that must be weighed, including but not limited to cost, noise, performance and risk.

But who ever said you actually had to decide one way or the other? Why not do both?

James from Major Hardware recently conceived this very idea and acted upon it. Using a Cooler Master Hyper 212 EVO as the base, James built an acrylic enclosure around the heatsink, attached some fittings and connected tubing. After a quick leak check and applying some extra silicon, the prototype watercooled heatsink was ready for a test run.

There are plenty of reasons why this wouldn’t’ be feasible over the long-term – corrosion and leaks immediately come to mind. The bigger question, however, is just how practical a setup like this is.

What are your thoughts? Could this catch on as a trend or is it little more than a fun YouTube experiment?

Permalink to story.

 
James, great vid. Thanks.

I would love to see an external water cooled (bathtub full?) using a chopped 212 EVO as the 'waterblock'.
 
There isn't much outside-the-box thinking here, as such approach has been used for many decades. Primarily, for the nuclear-reactor cooling systems.

b.t.w., use of water in this instance is very inefficient, as it naturally crystallizes on approach to zero. There are many types of liquid one can purchase cheap that can be cooled to well below zero, and thus offer much better efficiency. For example, ethanol freezes at -114C / −173 °F.
 
If it breaks the MB is guaranteed dead.
on the other hand, I had small leaks in my custom loop,
but all components for the most part were not under the water connectors and reservoirs.
 
Edited I believe this idea is in the spirit of making a more cost efficient system, not to make the best water cooling system ever.

p.s. Mmm he really needs to make the water case encompass the very bottom where the metal directly contacts the top of the base of the heat sink which is pasted to the cpu and lay the whole box flat. Being flat would make it closer to the cpu drawing out more heat, increase water circulation and allow the circulated water to hit more pieces of the heat sink. I suck at drawing 1 moment.

https://ibb.co/4FphCRJ
Black is the cpu, green is the heat sink bottom plate, yellow extended heat sink etc, I also changed the hose placement for a better flow.

p.s.s. since there is no fan involved it would be nice to have the cpu on the other side of the motherboard so the cooling system doesn't get it in the way and the weight of the heatsinked tank would be hunkered down by the mother board screws or whatever they are called. You could make the heat sinked tank pretty god damn big if you wanted too.

p.s.s.s. ok ok ok you could use a fan\radiator instead of a pitcher of water or you could use a 30$ water sealed mini fridge? or mini freezer\ethanol

psx 1 million actually id use both the fridge and a radiator

https://www.ebay.com/itm/Igloo-Mini-Compact-Refrigerator-Red/312511144979?_trkparms=aid=555018&algo=PL.SIM&ao=1&asc=20131003132420&meid=4d9d214f63664a25a0e55383a44778f5&pid=100005&rk=1&rkt=12&mehot=pp&sd=323512898354&itm=312511144979&_trksid=p2047675.c100005.m1851

If they made mobos with the cpus on the other side..... You wouldn't even have to worry about leaks. ignore the purple, red is the mobo, pink is the video card etc, black cpu, green waterblock, yeller heatsink
https://ibb.co/8MVwYLB

I looked it up, they say you can put the radiator in a mini fridge without a fan on it, but you risk condensation when your computer is off.
 
Last edited:
I can regard this just as a proof of concept,I.e. that everyone can make a water cooling solution
made out of inexpensive parts .Otherwise ,its technically irrelevant and pretty much inferior to a
proper commercial water block .Aluminum fins are absolute waste of space here as you need to cool only
the copper pipes,so it can be a much smaller package, especially with some other heat pipe cooler with
smaller footprint .As for inefficiency , with this you cool a copper filled with a fluid - usually water,instead
of circulating the water inside a copper ...
 
Ok now I am obsessed with tinkering around with experimental water cooling.... here are some cooool ideas
https://www.overclock.net/forum/134...-pc-no-but-what-about-radiator-freezer-9.html

I was thinking of setting up a fan radiator leading into a string of usb mini fridges (like 10$ each) and putting copper piping through the mini fridges. They would turn off with your comput, so no condensation probs and if they burn out meh. though I gonna go pick up some old thrift shop clunker pc to play with
 
There isn't much outside-the-box thinking here, as such approach has been used for many decades. Primarily, for the nuclear-reactor cooling systems.

b.t.w., use of water in this instance is very inefficient, as it naturally crystallizes on approach to zero. There are many types of liquid one can purchase cheap that can be cooled to well below zero, and thus offer much better efficiency. For example, ethanol freezes at -114C / −173 °F.
I believe he was only using ICE because he didnt have a radiator. You NEVER want to use any other liquid for cooling in instances such as this. water is always the most efficient because of its ultra-high specific heat and high boiling point. Also for a few bucks more you can purchase pure water instead of the tap and its non-conductive.
 
This cooler literally makes no sense in the long run. Why? Because in the end the cooling effectivity is determined how much heat the cooler can dissipate to the environment with virtually unlimited heat uptake capacity, which in this case is the air. However, this cooler greatly reduces both the heat exchange surface and also the heat exchange rate, because the tank has a smaller surface than all the radiator fins together, and because plastic conducts heat worse than does metal. So, all it does is worsen the cooling capability of the original cooler, with no added benefit whatsoever. It only works apparently better until the heat capacity of the water (and possibly) is exhausted, but from then on it will work even worse than a simple air cooled fan.
 
This cooler literally makes no sense in the long run. Why? Because in the end the cooling effectivity is determined how much heat the cooler can dissipate to the environment with virtually unlimited heat uptake capacity, which in this case is the air. However, this cooler greatly reduces both the heat exchange surface and also the heat exchange rate, because the tank has a smaller surface than all the radiator fins together, and because plastic conducts heat worse than does metal. So, all it does is worsen the cooling capability of the original cooler, with no added benefit whatsoever. It only works apparently better until the heat capacity of the water (and possibly) is exhausted, but from then on it will work even worse than a simple air cooled fan.
Seems some of you guys dont get out much. It was only for the sheer enjoyment of doing something people might of thought of doing but was silly overall. ie He built something for FUN.
 
Fine, but for the record, more correct logic imo, would be to start with the killer benefit liquid coolers have ... they transfer heat OUTSIDE the case - and THEN dissipate heat.

An air cooler is inevitably raising the ambient temp inside the case.

So it would be better to have a cheap liquid cooler with a copper cpu block, get the heat outside, and then improvise extra cooling for the budget radiator.
 
Back