What about using oil instead of water ???

G

game14

[H]ard|Gawd
Joined
Nov 8, 2006
Messages
1,454
Has anybody ever thought about using oil instead of water or actually done that ???
No i am not talking about filling up your rig with cooking oil !!!
I mean just using something like high quality synthetic motoroil instead of water in a regular watercooling setup .
Anybody ever used something similar ???
:D
 
Never done it, but I would think it would burn up your pump, as the viscosity is substansially higher than water.
 
There's no obvious advantage to that. Water carries heat perfectly well, its nearly free, safe to handle or transport and a mature technology.

Why would you want to use oil? It would be more expensive, not as safe to handle and a really difficult clean up.
 
One advantage in using oil would be that it has no conductivity at all .
So no problem with shortcut when the system leaks .
Another advantage is no corrosion of of the coolingparts , no matter what metal combination the circuit is build of , you could mix copper and aluminum or whatever you want .
About the viscosity I guess that very low viscosity Oil like in high performance Motoroil would be almost the same as the viscosity of non conductive Coolerfluids like Fluid XP or PC Ice .
The price for oil is the same as for high quality non conductive coolant .
About the cooling capacity of oil i dont know , i thought maybe someone could answer me that .
Tygon tubing should generally be safe for oilcooling , matter is what about the plastic used in the blocks ???

So it would be nice to hear some more opinions about my idea . :D
 
game14 said:
One advantage in using oil would be that it has no conductivity at all .
So no problem with shortcut when the system leaks .
Another advantage is no corrosion of of the coolingparts , no matter what metal combination the circuit is build of , you could mix copper and aluminum or whatever you want .
About the viscosity I guess that very low viscosity Oil like in high performance Motoroil would be almost the same as the viscosity of non conductive Coolerfluids like Fluid XP or PC Ice .
The price for oil is the same as for high quality non conductive coolant .
About the cooling capacity of oil i dont know , i thought maybe someone could answer me that .
Tygon tubing should generally be safe for oilcooling , matter is what about the plastic used in the blocks ???

So it would be nice to hear some more opinions about my idea . :D
Click to expand...

Many motor oils actually are conductive. Viscosity of even a 5w oil is significantly higher than water. Combine lower flow rates with lower heat carrying capacity and a really difficult cleanup when it spills. Also, you've got to re-engineer the materials used in the system so the oil doesn't break them down.

Those are some pretty solid and substantial reasons not to do it. I'm positive its technically possible, but there's so many disadvantages compared to water I can't see why you'd want to.
 
The reason why i came up with this idea is because of a video i have seen on tomshardware.com .
They filled a complete case with all its parts in it with cooking oil and it actually worked .
I mean there is no way i would do such thing like filling up my case with cooking oil , anyways cooking oil wouldnt last very long before it starts to smell .
I have also heard about a guy who used babyoil to cool his PSU , seems it worked .
So my thought was , why to do such a mess to bath all parts in oil , why not simply use oil instead of water in a regular waterloop .
Thought maybe someone would have already done that .... i mean u never know !!! :D
 
I've heard about the video your talking about, and in fact, there is someone here on [H] that did the same thing, basically anyhow.

I'm pretty sure these things were done simply to see "what if??"
I can almost guarentee you using motor oil would eventually cause problems, and I'm sure it would ultimately burn up your pump.
 
You could always go down to autozone and buy a pump designed for motor oil. :)
 
This is certainly something you could try just for the sake of it, the 'what if' factor. I'm a big advocate of that!

But think about it, car engines use oil for lubrication to stop wear between metal and metal. What do they use for cooling the engine?...you guessed it, water and additives - it's so much more effective.
 
cooking oil as long as it's kept warm like 100'ish *F will be quite liquid how ever not as much as water..

soy based stuff should work pretty well it's pretty clear when new too and comes in a liquid form aswell as a solid (both white or clear when cool and nearly water clear when at high temps

not sure on the corosivenes of it.. but it's not petrolium based so shouldn't eat any plastic's or rubber's haha rubber's

and you could always purchase a car fuel pump thats designed to be submerged in the tank. it can produce some high pressure so it will pump the oil..
 
Mysterae said:
This is certainly something you could try just for the sake of it, the 'what if' factor. I'm a big advocate of that!

Yeah maybe if i find an old computer for free , so i dont have worry if the "What if" factor goes wrong .

But think about it, car engines use oil for lubrication to stop wear between metal and metal. What do they use for cooling the engine?...you guessed it, water and additives - it's so much more effective.
Click to expand...

In certain point ur right , but when it comes to cool a turbocharger many times an oilcooler is used to cool it down . I guess they use water to cool the engine because its simply easier to use and u can keep the presure of the coolingsystem down .
I guess a proper working oilcooled system would mean to work with higher pressure , may cause the blocks to leak .
On the other hand it doesnt have to be exactly motoroil ...there a lots of oils out there that might work .
 
omegatotal said:
cooking oil as long as it's kept warm like 100'ish *F will be quite liquid how ever not as much as water..

soy based stuff should work pretty well it's pretty clear when new too and comes in a liquid form aswell as a solid (both white or clear when cool and nearly water clear when at high temps

not sure on the corosivenes of it.. but it's not petrolium based so shouldn't eat any plastic's or rubber's haha rubber's

and you could always purchase a car fuel pump thats designed to be submerged in the tank. it can produce some high pressure so it will pump the oil..
Click to expand...

Sounds great ...once u heat ur system up u never let it cool down ....if u do u would have fat instead of oil in your loop . Quite funny . :D
 
digital_exhaust said:
I've heard about the video your talking about, and in fact, there is someone here on [H] that did the same thing, basically anyhow.
Click to expand...

:D
------
The specific heat capacity of water is much better than hydrocarbon oils.If you were to put oil in a waterloop , it would probably run with a 50-75% reduction in efficiency.

Also ,There's actually quite a large difference between a submersion cooling setup such as mine or the tom's thing , and a waterloop . A waterloop cools the components by mechanical contact with a small area ; it needs to be efficient.Submersion cooling utilizes a much larger percentage of the surface area of the device , and can thus be less efficient.
 
if you had actualy read the post with that video on toms hardware, you would read that they used a special, non conductive, oil enginered by dupont at i believe 90 something dollars per gallon. most vegitable oils along with motor oil ARE electricly conductive, even more so then water.

http://www.museumofhoaxes.com/hoax/weblog/comments/3153/P20/

here is another article, there are hundreds of these stories out there, note the URL
museum of hoaxes dot com
not saying its impossible, it has been done, just dont think you can use any vegitable oil and have a super computer.
 
BrainEater said:
:D
------
The specific heat capacity of water is much better than hydrocarbon oils.If you were to put oil in a waterloop , it would probably run with a 50-75% reduction in efficiency.

Also ,There's actually quite a large difference between a submersion cooling setup such as mine or the tom's thing , and a waterloop . A waterloop cools the components by mechanical contact with a small area ; it needs to be efficient.Submersion cooling utilizes a much larger percentage of the surface area of the device , and can thus be less efficient.
Click to expand...

Thanks very much for the link , that page is great !!!
So AMMONIA would be the best to cool , about 25-50% better than water .
Man that really stinks :D
 
Ammonia only seems to have higher specific heats at higher temps. probably in the 10-20% better at normal op temps, and you'd still have to redesign everything to deal with the different viscocity etc.
 
ghost6303 said:
if you had actualy read the post with that video on toms hardware, you would read that they used a special, non conductive, oil enginered by dupont at i believe 90 something dollars per gallon. most vegitable oils along with motor oil ARE electricly conductive, even more so then water.

http://www.museumofhoaxes.com/hoax/weblog/comments/3153/P20/
Click to expand...

Thanks for the link , but one question , if oils are conductive how can you bath a whole system in it and it still works , guess nobody would try that with water !!!
About the video at toms hardware , they used 100 % seed flax oil u can buy at any grocery store in Europe . Its actually one of the cheapest oils u can get . U can see it clearly in the video ...maybe i find the link someday .
On the other side there is no way u could use vegetable oils for cooling over a long time , they brake down fast and would clock the whole system and smell really bad .
But thanks for the link and ur opinion. ;)
 
medel said:
Ammonia only seems to have higher specific heats at higher temps. probably in the 10-20% better at normal op temps, and you'd still have to redesign everything to deal with the different viscocity etc.
Click to expand...

And probably a huge fan to ventilate the room , which obviosly means a lot of noise !!! :D
 
game14 said:
In certain point ur right , but when it comes to cool a turbocharger many times an oilcooler is used to cool it down .
Click to expand...
Could this be because Oil has a higher boiling point than water? I could imagine that steam could cause a cooling system failure? Also, the oil could be used as a dual-agent: cooling and lubricating at the same time, making the design simpler.
 
game14 said:
And probably a huge fan to ventilate the room , which obviosly means a lot of noise !!! :D
Click to expand...

Well the old fashioned (ww1 era) fridges used amonia as a working fluid. If you design it right you don't have to worry about gas in the air during normal operation, but if it breaks overnight you're dead.
 
DanNeely said:
Well the old fashioned (ww1 era) fridges used amonia as a working fluid. If you design it right you don't have to worry about gas in the air during normal operation, but if it breaks overnight you're dead.
Click to expand...

thats sounds interesting ....what are they using now as working fluid ???
Or maybe using water or oil in an internal loop to cool the blocks and using an external heatexchanger system filled with amonia to cool it down .
What size is the smallest fridge ??? :D
 
R134a, the same HCFC that's used in your AC. Older ones would use Freon (R12).

* HydroChloroFlouroCarbons, just like the original, only not as ozone hungry. :D
 
I would think a system using R134a would be quite a bit more expensive. R134a is designed such that it has a phase change during operation (Google "refrigeration cycle", specifically vapor compression, if you want to learn about it).

Also the thought of using oil to cool a turbo in an automotive application is correct but the way you are thinking about it is wrong. Oil is used to lubricate the shaft in the turbo. The shaft spins at very high rpm and the oil heats up. The oil cooler is added to do just as it says, cool the oil. This in turn cools the turbo, but that is just a side effect of the goal of cooling the oil. If the oil gets too hot, it either has too low of viscosity to do any good lubricating the shaft or burns up.
 
game14 said:
In certain point ur right , but when it comes to cool a turbocharger many times an oilcooler is used to cool it down . I guess they use water to cool the engine because its simply easier to use and u can keep the presure of the coolingsystem down .
I guess a proper working oilcooled system would mean to work with higher pressure , may cause the blocks to leak .
On the other hand it doesnt have to be exactly motoroil ...there a lots of oils out there that might work .
Click to expand...
They cool a turbocharger with an oil cooler to keep the oil in the turbocharger cooler and keep it from breaking down. The oil is used primarily as a lubricant, not a coolant.
 
Just to address concerns about water's conductivity versus oil's, there are plenty of additives out on the market that reduce water's conductivity significantly so that even if there is a leak in the loop, it's a simple matter of cleanup and not of dead hardware.
 
Zoogle said:
Just to address concerns about water's conductivity versus oil's, there are plenty of additives out on the market that reduce water's conductivity significantly so that even if there is a leak in the loop, it's a simple matter of cleanup and not of dead hardware.
Click to expand...

NO!

They are not additives, they are coolants all their own. Do NOT, I repeat, do NOT combine a non-conductive fluid with water. That would take away the whole non-conductiveness of the fluid.

Water does not conduct electricity as long as it is just water. The problem with that is the second you put it in contact with computer parts it isn't just water anymore. It picks up all sorts of contaminants from everything it touches. These contaminants are what make it electrically conductive. Oil doesn't have this problem which is why it can be used for submersion cooling and water cannot.

And to the original question, as was mentioned many times water is a better coolant than oil, end of discussion.
 
Erasmus354 said:
NO!

They are not additives, they are coolants all their own. Do NOT, I repeat, do NOT combine a non-conductive fluid with water. That would take away the whole non-conductiveness of the fluid.

Water does not conduct electricity as long as it is just water. The problem with that is the second you put it in contact with computer parts it isn't just water anymore. It picks up all sorts of contaminants from everything it touches. These contaminants are what make it electrically conductive. Oil doesn't have this problem which is why it can be used for submersion cooling and water cannot.

And to the original question, as was mentioned many times water is a better coolant than oil, end of discussion.
Click to expand...

I have to agree here. Can it be done, sure! Is it worth the expense and trouble, NO!

You got to remember, The "Water" coming out of your tap, is not Pure water! Pure water does not conduct electricity, Pure water will also mix with Oil, Problem is it is near impossible to get/make Pure water, Tap water as well as any other common form including Distelled and Reverse Osmosis water contains minerals, gases, and other crap that makes it conductive.

Story on mixing Water and Oil
 
Sorry I lied :). Deionized water is good. Pick it up at your local chemistry department.
 
Zoogle said:
Sorry I lied :). Deionized water is good. Pick it up at your local chemistry department.
Click to expand...

Deionized / Distilled water doesn't change what I said. As soon as it comes in contact with the metals in your loop it will start to pick up contaminants. Water tends to do that with just about everything it comes in contact with. It will not stay non-conductive for long.
 
Altho proper pure deionized will collect contaminants (and ions) much quicker than distilled and actually cause a corrosive nightmare in most systems (if memory serves correctly). Lab-Grade de-ionized water is VERY corrosive. Distilled is the preferred variant for our purposes.
 
drivn2fast said:
I would think a system using R134a would be quite a bit more expensive. R134a is designed such that it has a phase change during operation (Google "refrigeration cycle", specifically vapor compression, if you want to learn about it).

Also the thought of using oil to cool a turbo in an automotive application is correct but the way you are thinking about it is wrong. Oil is used to lubricate the shaft in the turbo. The shaft spins at very high rpm and the oil heats up. The oil cooler is added to do just as it says, cool the oil. This in turn cools the turbo, but that is just a side effect of the goal of cooling the oil. If the oil gets too hot, it either has too low of viscosity to do any good lubricating the shaft or burns up.
Click to expand...

Thanks man , i asked my mechanic and he confirmed what u said .
About using a heatexchanger to cool a computer its already be done , i found it under extrem cooling solutions here . :D
 
Marci said:
Altho proper pure deionized will collect contaminants (and ions) much quicker than distilled and actually cause a corrosive nightmare in most systems (if memory serves correctly). Lab-Grade de-ionized water is VERY corrosive. Distilled is the preferred variant for our purposes.
Click to expand...

Anyways whatever water u use its becomes contaminated fast and so more conductive . On the other side its impossible to build a cooling loop without using more than one metal , so with using water as coolant u cant avoid corrosion and thats definitly a big con .
There must be any cooling liquid outthere that cools as good as water and is not conductive and keeps corrosion to a minimum . :confused:
 
Diablo2K said:
I have to agree here. Can it be done, sure! Is it worth the expense and trouble, NO!

You got to remember, The "Water" coming out of your tap, is not Pure water! Pure water does not conduct electricity, Pure water will also mix with Oil, Problem is it is near impossible to get/make Pure water, Tap water as well as any other common form including Distelled and Reverse Osmosis water contains minerals, gases, and other crap that makes it conductive.

Story on mixing Water and Oil
Click to expand...

that story about mixing oil with water is very interesting , but i guess not very easy to do at home :D
 
Antifreeze used for cars is conductive and has a higher viscosity and doesnt smell to good . :D
 
Wow... Lot of misinformation floating around in this thread.

As far as oil conductivity goes, I know mineral oil will withstand tens of thousands of kilovolts. Everything conducts electricity to some degree, even air. But at what point does it become a problem?
 
You must log in or register to reply here.
Back
Top