Slashdot Mirror
Aquarium Full of Oil For PC Cooling
JaredOfEuropa writes "Forget fancy watercooled CPUs or complicated heat pipes. Annoyed with the noise of the forced-air cooling in his computer, this guy simply dumped his entire motherboard in an aquarium filled with mineral oil. (coral cache). No modifications were necessary; he even left the fans running to keep the oil moving about. The only thing not submersed in oil is the hard disk."
This has been done before, as reported by slashdot almost six years ago. Of course, the guy in the '99 story used a styrofoam cooler, while the newer one upgraded to an aquarium, so I guess progress marches on!
Didnt some guy do this last year, and overclocked his 2ghz P4 to 4ghz? i wonder how fast his cd rom drives are submerged in mineral oil ;)
Mineral Oil is not nice stuff
Did you see the parts about flammable and a respiratory hazard?
What's next? A guy who uses gasoline for liquid cooling?
May I recommend Fluorinert FC-70?
"I'd rather be a lightning rod than a seismometer." -Ken Kesey
Isn't a mirror kind of useless if it has to pull the images from the original site's server? Just askin'
Give a man fire, and you warm him for the night. Set a man on fire, and you warm him for the rest of his life.
Attention all karma whores: please visit this previous story to obtain highly rated comments to be reposted here. That is all.
Why wouldn't they? Oil doesn't conduct. That's why it's used for cooling in electrical devices such as transformers, dummy-loads and such. I recall one vendor who demonstrated the high breakdown voltage of their oil by running a TV set in a vat of the stuff. Almost anything has better heat removal ability than air and for silent running it's not a bad idea.
There is still the problem of removing the heat. If there is enough surface area to allow the heat to be removed then you are ok, otherwise the oil (and everything else) will get too hot. Encasing everything in a metal box with fins on the outside would probably keep things even cooler.
~~~~~~~
"You are not remembered for doing what is expected of you." - Atul Chitnis
Except that the fans are still on, which supposedly moves the oil around. And when oil at the surface becomes warmer than room temperature, the heat is disappated into the air, with much greater surface area than is touching the heat source.
___
Cogito cogito, ergo cogito sum.
Yes, you can use water. Any non-conductive liquid will do. Of course tap water is conductive. you would have to use a very pure distilled water.
The oil will certainly be more difficult on the fans, im not sure if it has any corrosive effect.
No, you can't use water because it will become conductive when almost any substance is added to it, such as the metal flakes coming off of computer parts, other microscopic contaminents, and even carbon dioxide which will cause the pH to change. On the other hand, mineral oil is a non-conductive liquid, there are several although mineral oil is definatly the cheapest. Also, water is corrosive whereas mineral oil is not. A while back there was an article about using 3M Novec 1230 (AKA Sapphire) for submerged cooling (the poster mentioned it but forgot to take into account the low energy needed for Novec 1230 to vaporize) read the comments for more info on why water will not work. FYI somewhere in there you can find a link to pics of a guy who tried using water and fried his UPS.
The guy uses vegetable oil not mineral oil according to his site.
I did this for a year or so using mineral oil, a plastic storage tub and a small dorm-sized fridge. I had a small electric pump that pumped mineral oil into tubing which was coiled inside the fridge (drilled and in and out hole on the side) and then back into the resevoir. I was a little worried about condensation but it ran fine for over a year before I got tired of the clutter and mess of it. I could have done it better but I didn't want to spend any money on it and just use what I had laying around.
It was mostly for fun with a few interesting things I learned from it:
* It allowed me to overclock about 30% more than I could previously squeeze out.
* The mineral oil did not harm the hardware at all that I can tell from a year of being submerged(it just was a pain to clean).
* If you have your resevoir higher than your mouse then your mouse will be full of oil in a few weeks (same goes for any component connected by wire I imagine).
* The only component I found that could not be submerged was a hard drive.
* The outside coating on the wires will harden and break away after being submerged long enough(but they will still work).
* There was no connection issues with PCI cards or any peripheral device that was plugged in even if they were coated in mineral oil(even jumpers could be changed while it was submerged).
* If a drop of some other liquid (that is lighter than the oil) accidently falls into the resevoir it will quickly be coated by the mineral oil and slowly fall to the bottom and can be sucked out (phew!)
Probably more but those were the most interesting things I remember of it.
pretty much. here`s the mirrordot.
Stop Computers/Cars Analogies on S
don't give up your day job for a career in heat transfer...
A large surface area (e.g., the sides and top of the acquarium) makes an efficient device for convective heat transfer to the room air. The mineral oil would certainly get warmer than the room air, but it would stay well below the temp of the PC components.
Given sufficient motivation, the steady state value can be calculated within a fraction of a degree. Lacking that motivation, however, a reasonable approximation is that the acquarium would be less than 10C above room temp.
The electronics on some deep sea submarines are encased in mineral oil. Though immersing electronics in oil might sound strange, it is not an uncommon practice. Here are some research papers on the topic:
http://tdei.sju.edu/tdei/index/eiidx.html
Assuming you meant water, the you're wrong. Just water is quite a good insulator.
The US Navy was using water cooled computers long ago. Just flooded all the circuit cards with distilled water.
Note that you have to do this with pure water. If you dissolve much of anything in it, then the SOLUTION begins to conduct.
Most of the water you come into daily contact with (puddles, rivers, flooded basements, even tap water) has quite a bit dissovled into it, which is why generally electricity and "water" don't mix.
Someone did this a few years ago, but they pumped the oil over an air conditoner coil to get it even cooler, pics are awesome!
s ubmersion/submersion8.html
http://www.octools.com/index.cgi?caller=articles/
wouldnt there be problems caused on the mobo due to the changes in capacitance between all the conductive tracks?
No, because the trace impedance is set by the dialectric between the layers in the motherboard - it's the dialectric constant of the PCB material combined with the spacing between the trace and the plane beneath it, along with the trace width. Whatever is above the trace, in terms of what would normally be free air, makes virtually no difference, particularly since the motherboards already have a conformal coating with a fixed dialectric constant anyway.
But there are probably plenty of other reasons why vegetable oil isn't so great for your computer.
-h-
However, This site suggests a 75 watt heater to keep a 20 gallon tank 18 degrees F above room temperature, or a 150 watt heater for 36 degrees F.
The computer probably puts out less than 150 watts total. Even assuming an 80 degree F room, that would put the computer at 116 degrees F, which wouldn't upset the computer at all. Granted, the heater you put in a fish tank has a thermostat, and so it's not on all the time, but your computer will not have any problems at 116 degrees F inside, and could go a good deal higher safely.
But I do agree with the other guy to respond to your post -- I don't see the fans even turning, let alone turning enough to move the oil around. Perhaps if they were cut down some ...
Of course, I have no idea how well heat flows through oil, or how well it's transferred from oil to the air. But I imagine that the heat generated is low enough for it to not be a problem.
While the aquarium may have less total area, the conduction of heat to the aquarium walls through oil is much, much more efficient than through air to the walls of an aluminum case, resulting in far better heat dissipation overall.
Anyone who loves or hates any language, platform, or manufacturer, doesn't know what they're talking about.
as that high pressure hot air expands adiabatcally (no heat flow), it 'cools'. The added complexity of the compressor is not needed.
Its that density that has alot to do with it: more mass for it to offload the heat into per area vs. the air. Even if the fans were left off, oil is a much more effecient conductor of heat than air. To get straight to the point, the thermal conductivity (ie: how well it conducts heat) of air is 0.024Watts per Meter*degree Kelvin, vs that of Oil (machine oil in this case, but mineral will be on the same order of magnitude) at 0.15W/MK. Water would be better if it didn't tend to let the electrons go wherever they wanted, its conductivity is 0.58. The area of hot surfaces on the computer that are exposed is the same, since this is total emersion, and so long as the oil is moving enough to distribute the heat, the ammount of oil in the container is enough to serve as a decent heat sink, and the large surface areas of the top of the oil and sides of the aquarium would be sufficient for distributing that heat for the air to convect away.
How much heat this would work for would require thermo equations on the surface area of the exposed tank surfaces using convection (q=hA(dt)). h is the thermal transfer coefficient, and depends on velocity, density, geometry, flow pattern and a few other things, and since Im lazy Ill leave that as an exercize to the reader. Once you find h though (and for a flat plate like the aquarium walls and oil/air surface is, it should be easy), calculating the saturation point is simple.
been too long since I touched thermo...
tm
Support TBI Research: http://www.raisinhope.org
Somebody said it works.. but I think it's just from speculation. No real voltage checks or anything. Just that the computer accepted it.
Could be fine, could be sketchy. I ran my sound card 30 percent un-mounted for months before I realized why my computer locked up once in a blue moon.
Hermetically sealed, yes, airtight, no. They have to be able to let pressure vent for altitude change. Also, if it had a tiny leak.... Hard Disk Drives need air to work, the spinning motion of the platter creates a cushion of air that the read/write head rides on. If something got on the platter you can bet the drive will be destroyed.
Nope. They have small holes to let air move in and out. There's a filter behind the hole. There are often bigger holes behind warranty-void-if-removed stickers.
The heads are design to fly over the platters on an air bearing so disc drives don't work in either a vacuum or in oil.
But, the holes are small and oil viscous. It might run for a while. Maybe someone with a PC-in-a-tank and an old hard drive could experiement to see how long one lasts.
Iz
Actually, if you change the permitivity of the layer ABOVE the trace, you change the the "effective" permitivity of the transmission line (since e_r of oil is greater than e_r of air, you will increase the effective e_r of the transmission line). This will cause a net change in impedance of the transmission line. This cause cause havoc with timing. (check out "Microwave Engineering", Pozar) Just a thought.
size weight and cost
for any non-trivial power dissipation a heatsink with a fan is a far better bet on all of those criteria than a passive heatsink
while some heat leaves a heatsink by radiation a lot goes by conduction into the surrounding air. Whilst that air will move away by convection to some extent this isn't exactly a fast process especially in confined spaces like a computer case.
fans make the airflow both significantly faster and more predicatable.
the only real problem with fan cooling in such applications is noise.
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
Radiative cooling does not work particularly well until something like 250C, few (if any) semiconductors would ever survive using this cooling method... most power semiconductors are specified for 125-175C maximum junction termperature while most logic devices fall in the 65-75C range.
As for fan noise, most HSF have this flaw in common: they place the fan immediately on the heatsink. Every time a blade crosses a fin, this messes up the airflow, generating noise and pressure losses which reduce effectiveness/airflow. Another common issue with typical HSF is that because the fan is directly on top, the heatsink's center is an air flow dead-zone.
I have not seen them for myself but Intel's orb-like P4 HSF and their BTX reference design seem like good examples of proper designs, they both provide some clearance and avoid dead-zones.
"Oil doesn't conduct"
:)
Wrong. Everything conducts. Just certain things have much higher inate resistance properties. Like Oil