Mineral Oil Aquarium PC Now Allows EATX Mobos

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
[citation][nom]universaltrooper[/nom]Fan-less PC - just add water! No, wait...Seriously, ten gallons of mineral oil in a PC sounds like a big fire hazard in my opinion. And what about the fumes from the oil?[/citation]

the flash point of mineral oil (the temp at which it catches fire) is over 500F, so it would be safer than a deep fryer in the kitchen.
 
3M makes several fluids that will not void your warranty and are non-toxic. If you are spending 670 you may as well spend an extra 2-300 and get a fluid designed for the application you intended.
 
only problem I see is if the hdds are submerged as well in the oil and the normal seal on the shell of the hdd failed and oil got into the hdd case itself, the oil being thicker than air might mess with the seek arm and/or motor spinning the disks themselves. Also what about additional friction on the fans from circulating the oil as opposed to air. I realize heat would not be the issue but wouldn't that also put additional stress on the fan motors causing them to fail prematurely. Im mainly thinking about the ones mounted to the graffics cards.
 
[citation][nom]loomis86[/nom]No such thing as pure water. even if you could create pure water, as soon as you place it in contact with metalic objects, it will become impure. It will absorb metalic ions and start conducting electricity. if there is electric potential, this process happens even faster.[/citation]

Actually, the way to get pure water is indeed to use electricity. Ever hear of an ionizer?
 
To biohazard420420 hard drives are definite no no as they are not sealed at all they have a so called breather hole that must remain open to atmosphere to allow the drive to compensate for elevation changes. The breather hole does have a sort of filter though to keep out any external contamination.
 
Happy times if the thing breaks and you end up cleaning 10 gallons of mineral oil from your floor. I think I'll pass.
 
[citation][nom]abel2[/nom]Uhh, you wanna explain how water is going to condense when there isn't any water to condense in the first place? If you submerge something cold as hell into a tank full of oil, there is no water in that oil, and therefore no water can condense on the cold object. I would learn some basic chemistry before I start bashing someone about their engineering potential there wotan. And as long as you kept the temp above freezing point of mineral oil.. Something like -30 degrees F then it should be fine.[/citation]
this wotan guy is a complete idiot. i see him post on numerous articles and every response he makes is just as idiotic and uneducated as the next.

may I add that he actually stated Windows ME was better than windows 7, and linux is the best operating system on the market. and with this all said, he so claims that hes been in the IT industry for 15 years. Wow. sure does sound like he knows what hes talking about. I'm 17 and i know more than this loser lol.
 
[citation][nom]azcoyote[/nom]$671.46? That had better come with a drain basket and a frozen bag of Jalapeño poppers.[/citation]

$670 is nothing for a decent rig............

We have some cheap people on this site.
 
[citation][nom]abel2[/nom]Uhh, you wanna explain how water is going to condense when there isn't any water to condense in the first place? If you submerge something cold as hell into a tank full of oil, there is no water in that oil, and therefore no water can condense on the cold object. I would learn some basic chemistry before I start bashing someone about their engineering potential there wotan. And as long as you kept the temp above freezing point of mineral oil.. Something like -30 degrees F then it should be fine.[/citation]
I kind of understand where he is coming from. The freon units are dipped in the oil, the heat is drawn into the freon units and the heat is disipated in the freon units radiators. Problem here is that the radiators on the freon units have to work much hard because they are trying to remove energy from a substance with a differant SHC (specific heat capacity) than water, a greater volume (typical watercoolers have a smaller internal reservoir that is then refilled from mains or bottle) and the oil will also be warm, instead of water that is room temp or from the mains.

So oil - more volume of it, more heat for the same volume and harder to remove the heat it does have even if the volume was the same. The radiator that would typically be attached to the other end of the freon unit would have to be the size of a aircraft carrier and a fan so powerful you could use it on a helicopter crane.
 
[citation][nom]haze4peace[/nom]If I owned my own PC shop I would build one of these for decoration.[/citation]
"PC shop"... that is sooo 90's :)
... funny thing is.. i was thinking the same thing 😉
 
Well the guys at that computer shop just discovered a really ineffective way to do something that has been going on with supercomputers for decades. Right? I guess they never bother to consider that the tank itself should be a huge heatsink. Preferably made out of diamonds but silver or copper will do also. Glass and plastic are terrible as far as thermal transfer. (check ebay for copper containers under copper:kettles, basins. boilers, pots, etc. You can usually pick something up for around the price of scrap which I believe is less than the spot market price of about $3 per pound. It's kind of funny you guys debating whether or not mineral oil is ok to use. Look out the window. See the transformer on the utility pole? Unless it's a new one and uses soy or sunflower oil it is full of mineral oil (also known as transformer oil--that should be a big hint) mineral oil has several advantages over "green" oils: no need for antioxidants and perservatives in the oil and no need for oxygen scavangers in the head space between the dielectric fluid and the lid. Also stuff won't grow in it. Mineral oil is also known as baby oil and the chances of it flashing because of the current used in a computer is nonexistant. Your system would overheat long before it reached the flash point of the oil (which I think should be around 350-400C. But the biggest plus of mineral oil is that it does not absorb atmospheric water. Also an insulating layer could easily be devised on the cheap that would float on the surface if you have active cooling of the oil. Also consider the factor that any water that condenses is functionally distilled water and unless it gets contaminated is itself dielectric and a much better thermal transfer agent than oil. In fact if you do develop a layer of water at the bottom (you won't) it may improve the performance of the system unless it becomes contaminated by the electronics that are submerged in it. Being that the parts will still be sporting a coat of oil even in a water layer I doubt they will rust. The idea of bubbling air thorough the oil seems like you would be increasing the insulating factor of the oil, because air is a great insulator. Instead of spending money on air bubblers like some mouthbreather you would be better served to buy some copper tubing and treat the inner wall by scoring it with grooves capping it while evacuating the atmosphere with wet steam. Cap the top first. You now have heat pipes. Run you heat pipes down through the lid and have them touch the hotest areas. This is much more efficient at thermal transfer than oil. That is the big downside. This side of a dielectric slurry with 20% nano diamond content there just isn't any dielectric fluid that has high thermal transference. And that would be expensive. Better to use your nano diamonds in your heat pipes. They will increase their efficienty by 50 to 100%. How to get them in there and maintain a vacuum I leave to you. Anyway I won't go on, because I could go on and on. There is a reason that military grade supercomputers had used mineral oil submersion with active cooling--it works, ask IBM. Maybe next time computer hobbyists rediscover the wheel they will use a circular design huh?
 
Mineral oil has one other major advantage over the synthetic dielectic fluids. It's relatively nontoxic, but don't rub it on you baby no matter what johnson and johnson say. It may increase the chances of skin cancer when exposed to sunlight because it increased uv absorbtion. Anyhow for those of you interested in mineral oil, you want the lightest viscosity avalible, the kind they give to horses is good. Here are some of the other aliases that mineral oil goes by, but remember it comes in different viscosities for different applications.

* adepsine oil
* alboline
* baby oil
* bayol 55
* cable oil
* bayol f
* blandlube
* blandol white mineral oil
* carnea 21
* clearteck
* crystol 325
* crystosol
* Diala-X, AX
* drakeol
* electrical insulating oil
* ervol
* filtrawhite
* fonoline
* frigol
* glymol
* Heat-treating oil
* hevyteck
* hydraulic oil
* hydrocarbon oils
* jute batching oil
* kaydol
* kondremul
* kremol
* LHM
* lignite oil
* liquid paraffin
* lubricating oil
* master Shimmer
* mineral oil (saturated parrafin oil)
* mineral oil hydrocarbon solvent (petroleum)
* mineral oil mist
* mineral oil, aromatic
* mineral oil, paraffinic
* mineral Seal Oil
* molol
* neo-cultol
* nujol
* oil mist
* oil mist, mineral, severely refined
* Oil mist, refined mineral
* oil, petroleum
* paraffin oil (class)
* paraffin oil
* parol
* paroleine
* peneteck
* penreco
* perfecta
* petrogalar
* petrolatum
* petroleum hydrocarbons
* petroleum, liquid
* primol
* primol 355
* primol d
* protopet
* saxol
* tech pet
* f triona b
* uvasol
* univolt N60, 80
* voltesso 35
* white mineral oil
* white oil
Just make sure they are pure (without additives) and the right thickness and buy the cheapest stuff. Actually some of the things used as additives would be ok. I just don't have the time to get into it.

So now all of a sudden you can build a much more efficient set up for way less than the money quoted in the article. Figure $60-$80 per five gallon container of oil and another hundred for the copper container and the copper tray you will buy to use for the lid and maybe fifty for incidentals. Right in the ballpark of medium priced computer case, but when you are finished with it, instead of a hunk of steal or aluminum scrap, you have an antique copper boiler (clean but do not shine it up) which you can probably sell on ebay and a bunch of oil that you can also sell or use in your vacuum pump (forgot that one--vacuum pump oil) or in your two stroke cycle or sell it to your local massage therapist.
Your welcome.
 
Buy the way for the persons who suggested using designer synthetic dielectric fluids: Most of them will not be any better than mineral oil and could present all kinds of additional problems. You can take my word for it or do the research. Sure it would be nice to have a swimming pool full of florinert. You could sit down there and compute without any scuba gear and breath the dielectric fluid without harm, but could you afford the $250 plus per gallon for something that is only marginally better at heat transfer than the $12 a gallon horse laxative you should have used. Remember what happened to those Aussies who spent the big bucks on Florinert and pumped it through a dry ice and acetone heat exchanger? It geled and they couldn't use it anyway. The main reason that alternative dielectrics are being developed is not because they transfer heat better, but because they degrade in the environment, and mineral oil, just like crude oil, hangs around a long time so it presents an environmental problem when it is drained or spills out of transformers. Since yours will be clean you won't have this problem so there is really no reason at all to pay more. And yes you could use vegetable oil. There have even been a number of studies done on what percentages of different veg oils are most effective, but these present all kinds of additional problems and will not stay clean, unless you get the professional stuff that is starting to be used in the electric grid, they are stupid expensive for what they are. You could develop you own additive system for veg oils but it's just not going to be as good and the viscosity will be higher and the thermal transfer not meaningfully better.
 
Here is a photo of a twenty pound plus solid copper vessel that recently sold on ebay for less than $50 dollars. I'm sure that you could buy something as good new for five or six hundred.

http://s36.photobucket.com/albums/e6/badboybob/?action=view&current=DSC04626.jpg

It has a hole near the bottom for a valve and is near the perfect size for any atx board. Holds 10 gallons. With 20lbs of copper, just attaching some short heatpipes between the hot component parts and the tank wall would probably cool it entirely passively were I live cause it almost never gets above 75 degrees. But some people would insist on running a copper cooling coil of some kind, either in the tank or pumping the oil out though an active system. By active I do not mean a oil to air heat exchanger. I mean, something like was mentioned with the water chillers or maybe with the coil submerged in an antifreeze filled gelato maker or other appropriate vapor phase devise. Bottom line, if you want to gaze into an aquarium fill it with fish. Computers are not for looking at, there's nothing to see. Unless you are very easily amused.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom