New Electronics Friendly Water

So I saw this the other day... a brand new mix of chemicals designed as a firefighting tool. The advantage of this substance is it doesn't "wet" objects. They dunked running laptops, LCD displays and books into it and pulled them out with no damage.

http://abclocal.go.com/wpvi/news/04132004_bb_sapphire.html

So I wonder if you could plunk a running computer in a tank of this stuff and just circulate and cool the water and not worry about tubes and cramming in a pump and all that.

Also, it would be great in standard cooling setups because if it leaked it would not damage your machine.

It's meant for use in sprinkler systems for museums, so I assume it will not be as cheap as water, it has to be fairly inexpensive for just a gallon or two.

The only question is it's heat transfer ability and how viscious it is. Even if it was not all that good, it has to be much better than air and would be a possible choice for a submurged cooling system.

I wonder how close to production it is.. would be fun to get some to play with and test in a system.

That stuff sounds pretty damn sweet. If it had good properties it would be a real winner. It can't be too expensive. Can you imagine how many gallons it would take to fill an entire sprinkler system of a building? One gal. couldnt' be too expensive.

It sounds like that it doesnt make a full contact with solids since books did not get wet... I mean if it actually makes a full contact then the book should be wet. May be it only has firefighting properties of water but nothing else...

Saw the article on /. looks interesting. Would have problems cooling things with it though wouldn't you? I will leave it up to the more knowledgeable to debate it.

http://www.fs-world.com/Show.asp?ID=3222&Form=5

From this link the stuff seems to be sorta like fire extinguisher stuff, I doubt it would work just in a tank with computer stuff like you would need.

I wouldn't think this stuff would work with a normal waterblock, but I'd think it be feasible for motherboard submersion (with actively cooling the fluid). Interesting indeed.

ILikeMy240sx said:

It sounds like that it doesnt make a full contact with solids since books did not get wet... I mean if it actually makes a full contact then the book should be wet. May be it only has firefighting properties of water but nothing else...

Click to expand...

It talks about "drying" the objects, could suggest absorbtion of such material...

maybe they mean the time it takes for the stuff to roll off thebook or something.

regardless, highly interesting

Seems to me like you are suggesting something like the chilled mineral oil cooled computers of early experiment overclocking era.... (I have not seem one of thost projects for AGES actually... )

Shadowcat said:

Seems to me like you are suggesting something like the chilled mineral oil cooled computers of early experiment overclocking era.... (I have not seem one of thost projects for AGES actually... )

Click to expand...

People still do them.

Someone should try to find out if there's a place where we can order some

can we use that in our watercooling units at least?? that way you wouldn't be worried at all if your system sprung a leak.

not to underestimate them, but in MOST cases(not all) but the same properties that let a substance conduct electricity let it conduct heat, they are the same thing, ceramic and oil is the only things that comes to mind as not.

so Id guess if you didnt have the advantage of 100% perfect surface contact on the components, it would be worse than water in a "normal WC setup".

Ok I looked into and found this:

Chemical Formula CF3CF2C(O)CF(CF3)2
Molecular Weight 316.04
Boiling Point @ 1 atm 49.2°C (120.6°F)
Freezing Point -108.0°C (-162.4°F)
Critical Temperature 168.7°C (335.6°F)
Critical Pressure 18.65 bar (270.44 psi)
Critical Volume 494.5 cc/mole (0.0251 ft3/lbm)
Critical Density 639.1 kg/m3 (39.91 lbm/ft3)
Density, Sat. Liquid 1.60 g/ml (99.9 lbm/ft3)
Density, Gas @ 1 atm 0.0136 g/ml (0.851 lbm/ft3)
Specific Volume, Gas @ 1 atm 0.0733 m3/kg (1.175 ft3/lb)
Specific Heat, Liquid 1.103 kJ/kg°C (0.2634 BTU/lb°F)
Specific Heat, Vapor @ 1 atm 0.891 kJ/kg°C (0.2127 BTU/lb°F)
Heat of Vaporization @ boiling point 88.0 kJ/kg (37.9 BTU/lb)
Liquid Viscosity @ 0°C/25°C 0.56/0.39 centistokes
Solubility of Water in Novec 1230 Fluid <0.001 % by wt.
Vapor Pressure 0.404 bar (5.85 psig)
Relative Dielectric Strength, 1 atm (N2=1.0) 2.3

This stuff boils at 49.2C and has a heat capacity of about 1/4 of water. Its 1.1 to waters 4.2

More or less I don't think this stuff would be that good for water-cooling. There are better coolants used for cascade systems.

VAdept said:

Ok I looked into and found this:

Chemical Formula CF3CF2C(O)CF(CF3)2
Molecular Weight 316.04
Boiling Point @ 1 atm 49.2°C (120.6°F)
Freezing Point -108.0°C (-162.4°F)
Critical Temperature 168.7°C (335.6°F)
Critical Pressure 18.65 bar (270.44 psi)
Critical Volume 494.5 cc/mole (0.0251 ft3/lbm)
Critical Density 639.1 kg/m3 (39.91 lbm/ft3)
Density, Sat. Liquid 1.60 g/ml (99.9 lbm/ft3)
Density, Gas @ 1 atm 0.0136 g/ml (0.851 lbm/ft3)
Specific Volume, Gas @ 1 atm 0.0733 m3/kg (1.175 ft3/lb)
Specific Heat, Liquid 1.103 kJ/kg°C (0.2634 BTU/lb°F)
Specific Heat, Vapor @ 1 atm 0.891 kJ/kg°C (0.2127 BTU/lb°F)
Heat of Vaporization @ boiling point 88.0 kJ/kg (37.9 BTU/lb)
Liquid Viscosity @ 0°C/25°C 0.56/0.39 centistokes
Solubility of Water in Novec 1230 Fluid <0.001 % by wt.
Vapor Pressure 0.404 bar (5.85 psig)
Relative Dielectric Strength, 1 atm (N2=1.0) 2.3

This stuff boils at 49.2C and has a heat capacity of about 1/4 of water. Its 1.1 to waters 4.2

More or less I don't think this stuff would be that good for water-cooling. There are better coolants used for cascade systems.

Click to expand...

Good stuff. What's exciting about that is the liquid viscosity at 0.39 centistokes at 25C. Water by comparison is around 0.88 centistokes at 25C.

At 0C, water is 1.79 centistokes, while this stuff is 0.56.

That makes this liquid about 2.5x more "runny" than water. That's actually pretty good because a low viscosity means a thinner boundary layer, which improves thermal convection when it's being pumped about.

It's Liquid specific heat is 1103 J/kg°C, and density of around 1.6 (water has a density of 1.0). This means that it holds around 1765J per litre-C, compared to water's ~4200J per litre-C, or around 42% of the thermal capacity of water by volume.

Generally speaking, one would want flow rates up above 3LPM with this stuff to keep its low thermal capacity from being a significant issue, but given its low viscosity that should be very easy to accomplish.

What we don't know about it is its thermal conductivity. Water is 0.6W/m-K. A thermal conductivity better than around 0.4W/m-K when coupled with this liquid's low viscosity, would make it better than water for forced (pumped) convection. The boiling point of 49.2C is not really much of an issue unless you run your computer in the middle of Death Valley or something.

Anyone know its thermal conductivity?

Cathar said:

Good stuff. What's exciting about that is the liquid viscosity at 0.39 centistokes at 25C. Water by comparison is around 0.88 centistokes at 25C.

At 0C, water is 1.79 centistokes, while this stuff is 0.56.

That makes this liquid about 2.5x more "runny" than water. That's actually pretty good because a low viscosity means a thinner boundary layer, which improves thermal convection when it's being pumped about.

It's Liquid specific heat is 1103 J/kg�C, and density of around 0.64 (water has a density of 1). This means that it holds around 706J per litre-C, compared to water's ~4200J per litre-C, or around 1/6th the thermal capacity of water by volume.

Generally speaking, one would want flow rates up around 10-12LPM with this stuff to keep its low thermal capacity from being a significant issue, but given its low viscosity that should be fairly easy to accomplish.

What we don't know about it is its thermal conductivity. Water is 0.6W/m-K. A thermal conductivity better than around 0.4W/m-K when coupled with this liquid's low viscosity, would make it better than water for forced (pumped) convection. The boiling point of 49.2C is not really much of an issue unless you run your computer in the middle of Death Valley or something.

Anyone know its thermal conductivity?

Click to expand...

it appears to be this stuff
3M Novec 1230 Fire Protection Fluid - i'm sure you could pry the answer out of them.....

It's different than the 3M stuff, but similar.

Hmmmm, it looked to me that it was the same as the 3M stuff, just that Tyco incorporated the 3M stuff into a whole fire system or whatever.
Anyways, anyone know where to order this 3M stuff?

Sterculus said:

Hmmmm, it looked to me that it was the same as the 3M stuff, just that Tyco incorporated the 3M stuff into a whole fire system or whatever.
Anyways, anyone know where to order this 3M stuff?

Click to expand...

http://cms.3m.com/cms/US/en/2-68/kRRrkFL/view.jhtml

Well, I sent them a message asking about buying some. We'll see what they say. If its reasonably priced, I may buy enough to do a full case immersion project sometime.

cather makes great points all around. but some thoughts... it doesnt need much thermal conductivity, cause unlike a water block where only small relatively minute amounts of liquid touches the hot sides of the block, this has 100% contact to the chip. with the viscosity you really could use air fan brought down to like 60RPMs to keep flow over the CPU, and keep things moving so it doesnt develop an internal current that bypasses the hot components.

be even cooler if you could set the cpu chip off the mount by like .1" or something(if the pins could reach. and thus pass the liquid over AND under the chip for maximum cooling.