Is mineral oil submerging a valid way for cooling?

[Garuger]

I dont know if it really belongs here but it is still a liquid cooling systems . I saw many builds of submerging their PC under mineral oil in fish tanks and i always wondered if it is still valid in 2014 and if it is better than watercooling because i have not seen any bench marks about it vs Aero cooling and water cooling and would be interesting to see how modern techonology would react to it .I know of course that the parts will be useless if they are removed . So anyone still does that in 2014 and it is better than watercooling? or is it just a fancy gimmick to impress ppl?

[RainColt]

I dont know if it really belongs here but it is still a liquid cooling systems . I saw many builds of submerging their PC under mineral oil in fish tanks and i always wondered if it is still valid in 2014 and if it is better than watercooling because i have not seen any bench marks about it vs Aero cooling and water cooling and would be interesting to see how modern techonology would react to it .I know of course that the parts will be useless if they are removed . So anyone still does that in 2014 and it is better than watercooling? or is it just a fancy gimmick to impress ppl?

It's more gimmick than it is efficient

[Garuger]

It's more gimmick than it is efficient

Are there benchmarks? 0_o

[AlwaysFSX]

It works but you'd have moderately higher temps.

[RainColt]

Are there benchmarks? 0_o

I've never seen any, I doubt that anyone is willing to trash good parts for the sake of a benchmark although a little birdy told me that Linus is planning an oil cooled build guide, maybe he'll do benchmarks (mind you that was like a month  or two ago)

[Garuger]

I've never seen any, I doubt that anyone is willing to trash good parts for the sake of a benchmark although a little birdy told me that Linus is planning an oil cooled build guide, maybe he'll do benchmarks (mind you that was like a month  or two ago)

Linus ? oiled cooling?? Let me guess 4 way sli gtx 980 with X99 chipset and 64 gb of ram ddr4??

[Garuger]

It works but you'd have moderately higher temps.

Well does not that depend on how fast you refresh your oil and the pump and other stuff?

[LarsReviews]

You are the first Greek I see here. ja su, ti kanis?

[Garuger]

You are the first Greek I see here. ja su, ti kanis?

Kala esy? kai autos ellinas einai http://linustechtips.com/main/user/126083-nekro/

[Swndlr]

Kala esy? kai autos ellinas einai http://linustechtips.com/main/user/126083-nekro/

dawwww forums bonding so cayuteeeeee

[RainColt]

Linus ? oiled cooling?? Let me guess 4 way sli gtx 980 with X99 chipset and 64 gb of ram ddr4??

No, I think he mentioned finding a 480 to use in his oilcooled build guide and another video where he drew attention to the oil.

[LarsReviews]

Kala esy? kai autos ellinas einai http://linustechtips.com/main/user/126083-nekro/

 

I am just particle greek, so after "how are you" my language skills end

[W-L]

I dont know if it really belongs here but it is still a liquid cooling systems . I saw many builds of submerging their PC under mineral oil in fish tanks and i always wondered if it is still valid in 2014 and if it is better than watercooling because i have not seen any bench marks about it vs Aero cooling and water cooling and would be interesting to see how modern techonology would react to it .I know of course that the parts will be useless if they are removed . So anyone still does that in 2014 and it is better than watercooling? or is it just a fancy gimmick to impress ppl?

 

From what I've heard it's more efficient to do a watercooled system with multiple radiators than oil cooling since mineral oil isn't the best at conducting heat away from concentrated sources, the benefit is that is cools everything that is submerged inside of it and not just solely the CPU or GPU, there was a build I saw of a watercooled loop that was submerged in mineral oil to perform the best in both situations whereby the main components are watercooled and are a separate system to the oil itself.

[Garuger]

From what I've heard it's more efficient to do a watercooled system with multiple radiators than oil cooling since mineral oil isn't the best at conducting heat away from concentrated sources, the benefit is that is cools everything that is submerged inside of it and not just solely the CPU or GPU, there was a build I saw of a watercooled loop that was submerged in mineral oil to perform the best in both situations whereby the main components are watercooled and are a separate system to the oil itself.

That is all fine and dandy but there are no benchmarks wich makes me kinda sad.It is like finding a rare animal and never record its performance 

[W-L]

That is all fine and dandy but there are no benchmarks wich makes me kinda sad.It is like finding a rare animal and never record its performance 

 

True it's be interesting to see real world comparisons

[Grrizz]

Mineral oil has a higher viscosity, lower specific heat capacity, lower thermal conductance, is quite pricey and the submerging of parts has several cravats you need to work around like cable wicking/siphoning, you only do it for looks and/or bragging rights as water trumps oil at every point except electrical conductivity and weight

In saying that I have a oil build planned but haven't had the time to get it started yet. It wont be a performance build just a 10W Celeron made pretty for the lounge

[smjpl]

Valid  = Yes

 

Practical = No

 

Yeah it is pretty much a gimmick. Its cool if you're into that thing but it kinda wrecks your parts. They will still work fine but horrible to work with. Just use an air cooler (because air can be just as good as water cooling for 95% of circumstances) unless you want to spend extra money to have something look cool.

[Garuger]

Mineral oil has a higher viscosity, lower specific heat capacity, lower thermal conductance, is quite pricey and the submerging of parts has several cravats you need to work around like cable wicking/siphoning, you only do it for looks and/or bragging rights as water trumps oil at every point except electrical conductivity and weight

In saying that I have a oil build planned but haven't had the time to get it started yet. It wont be a performance build just a 10W Celeron made pretty for the lounge

I see , well i never intended to do that but i was just always qurious because they were no benchmarks

[Garuger]

Valid  = Yes

 

Practical = No

 

Yeah it is pretty much a gimmick. Its cool if you're into that thing but it kinda wrecks your parts. They will still work fine but horrible to work with. Just use an air cooler (because air can be just as good as water cooling for 95% of circumstances) unless you want to spend extra money to have something look cool.

Air cooling just need filters and a fan controller along with a good airflow case ,but i would prefer that than water cooling . Just my opinion thought

[Grrizz]

I see , well i never intended to do that but i was just always qurious because they were no benchmarks

I think the reasons you don't see many benchmarks is because the builds are usually done with older expendable hardware knowing performance wont be as good as water but also because the setups themselves are quite different, things like air coolers are usually dumped in the oil instead of heat sinks designed with fluid in mind so its hard to make valid comparisons, the best way to do it would probably be filling a water-cooling loop with oil and even then I would speculate older water cooling blocks might perform better due to their generally less restrictive designs vs. modern ones. The other problem with benchmarking oil rigs built in the traditional way is the volume of oil used, it takes a long time for the components to reach thermal equilibrium so each bench would take a lot of time to complete which I imagine most people don't consider worth while.

[Garuger]

I think the reasons you don't see many benchmarks is because the builds are usually done with older expendable hardware knowing performance wont be as good as water but also because the setups themselves are quite different, things like air coolers are usually dumped in the oil instead of heat sinks designed with fluid in mind so its hard to make valid comparisons, the best way to do it would probably be filling a water-cooling loop with oil and even then I would speculate older water cooling blocks might perform better due to their generally less restrictive designs vs. modern ones. The other problem with benchmarking oil rigs built in the traditional way is the volume of oil used, it takes a long time for the components to reach thermal equilibrium so each bench would take a lot of time to complete which I imagine most people don't consider worth while.

Well that makes sense .So we only need a rich guy for a 4 way sli of 980 with ddr4 and i7 any volunteers?

[Garuger]

Mineral oil has a higher viscosity, lower specific heat capacity, lower thermal conductance, is quite pricey and the submerging of parts has several cravats you need to work around like cable wicking/siphoning, you only do it for looks and/or bragging rights as water trumps oil at every point except electrical conductivity and weight

In saying that I have a oil build planned but haven't had the time to get it started yet. It wont be a performance build just a 10W Celeron made pretty for the lounge

Well some use it for servers look here :

there is so much controvesy on this matter

[Grrizz]

Well some use it for servers look here : ...snip... there is so much controvesy on this matter

For a server farm the ability to move the heat outside the building is nice which fluids are good for and setup expense is not such a big factor as down time costs them a lot more. The reason oil can be popular with servers instead of water is because its non-conductive, this means you can just drop your rack into the vat (assuming good circulation has been taken into account already) without having to deal with setting up a custom water loop in each rack and having to deal with the risk of a leak causing downtime or damage. It also allows a longer response time in case of failure, if a pump goes down you will generally have backups/multiple pumps running so you still get decent flow, likewise on the radiator if one fan goes down there are still several others to pick up the slack and in the case of a catastrophic failure you have the volume of the oil in the vat able to buffer the heat giving time for a proper shutdown/on the fly repair whereas with air cooling you only have the heat sink which will heat up very quickly and with water-cooling in the case of a pump/flow failure you only have the mass of the block and water contained within it as a buffer. Basically for them is a compromise between best possible cooling and reliability.

[smjpl]

Air cooling just need filters and a fan controller along with a good airflow case ,but i would prefer that than water cooling . Just my opinion thought

 

I would argue you need intake filters anyway (whether water or air) and a fan controller isn't necessary. Just decent fans. Fan controller can allows you to better tweak idle vs power loads but again isn't necessary.

[Garuger]

I would argue you need intake filters anyway (whether water or air) and a fan controller isn't necessary. Just decent fans. Fan controller can allows you to better tweak idle vs power loads but again isn't necessary.

I would like to have a fan controller because the mobo i aim to buy does not have that many spots for fans and the fans are a bit noise .Also do you know what happens if you overpass the max watt for each channel fan?