Anyone tried TransformerOil as an Additive?

[j813]

Question exactly what the thread's title is.
I heard it helps cool things, but is it flammable @ the same time?
Pls put on some experiences if you have.

[Butcher]

Oil and water don't really mix...

[Joe]

doesnt transformer oil have PCB's in it? you know, toxic, cancerous, etc...

[pdf27]

Oil + Water + Shaking about = good approximation to cream cheese!
The reason transformers use it for cooling is that it is a good electrical insulator, so can be used in places where water would be dangerous. You could use it as your only cooling fluid, but it's probably only worth it if:
1) Parts of your system have to be electrically live, or
2) You have extremely fierce galvanic corrosion problems.

[Butcher]

Originally Posted by Joe

doesnt transformer oil have PCB's in it? you know, toxic, cancerous, etc...

Seems oils you can currently purchase tend not to contain PCBs, they certainly used to though.
I'm still not sure what use as an additive transformer oil would have as it has significantly lower heat capacity than water (water has pretty much the highest heat capacity of any substance known to man in fact). Plus it would just float on the water.

[j813]

Found out this idea from this thread http://forums.amdmb.com/showthread.p...5&pagenumber=2

Originally Posted by pdf27 The reason transformers use it for cooling is that it is a good electrical insulator, so can be used in places where water would be dangerous. You could use it as your only cooling fluid, but it's probably only worth it if:
1) Parts of your system have to be electrically live, or
2) You have extremely fierce galvanic corrosion problems.

Can electronic parts while on with power be submerged in this type of oil?

Thanks for the inputs!!

[pdf27]

Think so - I did look at that thread and it made no sense at all - Cp of oil is about 40% of that for water so that you'll get higher temperatures. The real problem is that mixing oil and water and agitating it violently (as in a pump) will result in it emulsifying and becoming a solid...

The whole point of transformer oil is that it can be used as a coolant at high voltages (typically up to 400,000 Volts). 240v is no problem at all, and some computers have been cooled by submerging them directly in oil.
http://www.eppenga.com/folder.php?id=21

[Gooserider]

Modern transformer oils do not have PCB's or other allegedly toxic ingredients. (the claims about PCB's weren't based on the best of science...) As a coolant in a standard PC WC system it would be lousy, lower heat capacity than water, and harder to pump.

However some of the radical cooling folks have been known to immerse their entire systems (except for the hard drives which DON'T like running on anything but air) in tanks of mineral oil, transformer oil would do the same thing. They then use phase change or other systems to cool the oil. Some radical overclocks can be gotten this way, and it saves the hassles of cooling individual components, but I personally doubt that it's worth it.

Gooserider

[UberBlue]

Is he thinking about imersion cooling?

[killernoodle]

Water soluable (sp?) Oil exists in many types of industrial cutting oils, we happen to have a small barrel of it in our garage. Ours is made by pennzoil and we use it mixed with water in our band saw. It is also pumped, so there is no threat to it turning into a solid.

I'm not sure if it conducts electricity or not, so oh well.

Has anyone ever thought about adding like copper dust into a loop? I know it would destroy any pump pretty quickly, but do you think cooling capacity would be greater? I have always wanted to know.

[Butcher]

Copper has a lower specific heat capacity than water so it would reduce coolant efficiency.

Just to clear up a common misconception:
For a waterblock or radiator you are looking for thermal conductivity. The best materials known are diamond, silver and copper in that order.
For a coolant you are looking for specific heat capacity. The best materials known are hydrogen, helium and water in that order.

Heat Capacities for some common materials:

Material   Specific Heat Capacity (J/(kg˚C))
Aluminum   900
Copper     390
Glass      840
Iron       450
Marble     860
Wood       1700
Water      4186
Mercury    140
Hydrogen    14304

Thermal conductivity for some common materials:

Material   Thermal Conductivity (W/m-K)
Silver     406.0 
Copper     385.0 
Brass      109.0 
Aluminum   205.0 
Steel      50.2 
Lead       34.7 
Mercury    8.3 
Glass      0.8 
Concrete   0.8 
Cork board 0.04 
Wood       0.12-0.04

Also I found a handy report on the specific heat capacity of transformer oil / water mixtures:

Concentration   Specific Heat Capacity (J/(kg˚C))
20              2319
25              2436
50              3022
75              3608
90              3959

Note, that's concentration of water, not oil - as the amount of water is increased the specific heat capacity increases. Pure water is more efficient than any mixture.

[pdf27]

Surely the specific heat capacity per unit volume is more important for this sort of thing than the /per unit mass? In this case hydrogen and helium would do rather worse, while something like mercury might do rather well.

[prb123]

So...in theory would compressed liquid hydrogen be the best coolant to use...assuming you overcame the crazy radiator/pump/wb/pressure, etc.

[pdf27]

Liquid Hydrogen would be great - boiling point is about 20K!!!

[Butcher]

Originally Posted by pdf27

Surely the specific heat capacity per unit volume is more important for this sort of thing than the /per unit mass? In this case hydrogen and helium would do rather worse, while something like mercury might do rather well.

Possibly, yes, here's the numbers adjusted for the relative densities:

Material   Heat Capacity (J/(cm³˚C))
Aluminum   2.43
Copper     3.4788
Iron       3.5433
Water      4.186
Mercury    0.7602
Hydrogen   0.001273056

Numbers are given for 20C, except hydrogen which is quoted at 0C.

Originally Posted by pdf27

Liquid Hydrogen would be great - boiling point is about 20K!!!

One leak and say goodbye to your house though.

[|kbn|]

"The best materials known are hydrogen, helium and water in that order."

what about helium? one leak and your vioce goes funny, but that would be about it...?
whats heliums specific heat capacity/volume?

[Butcher]

Helium's heat capacity is 0.000927 J/(cm³˚C). While it's almost twice as dense as hydrogen, it's specific heat capacity is nearly 3 times lower at 5193, not that much above water. Also liquid helium is very hard to obtain as it's boiling point is 4.126K.
On the other hand, as you say it's completely inert.

[Gooserider]

Butcher said most of it better than I could . The only thing I would add is that while Helium is chemically inert, it would still be dangerous as a coolant because a leak could easily displace the oxygen in the air. Helium may make you talk funny, but you can't live on it very well. (Unfortuneately some party animals have found this out the hard way, don't over do the stuff....)

Gooserider