Weird water cooling question

Toggle sidebar Toggle sidebar
M

Mnx4

Distinguished
Aug 10, 2001
324
0
18,780
I was thinking the other day how to make water cooling work better, like what liquid(s) could I use to make the heat transfer work better. My question is does anyone think that replacing water with a liquid like acetone or rubbing alcohol would make it work better? Both have a high vapor pressure and in turn, they cool off because they evaporate quickly. I know its crazy but could it work? Please give me some info! Thanks.

Do you work at UPS? 'cuz i could have sworn you were checkin out my package!
 
B

Boondock_Saint

Distinguished
Jul 4, 2001
548
0
18,980
I am not positive, but evaporation speeds will have nothing to do with how much heat it can hold/carry. The specific heat of acetone or rubbing alcohol might be much lower than water, making it much less effective. Remember this is a closed circuit, so no evaporation can occur. :smile:

Try checking into the specific heat of various compounds and substances to use in it.

:tongue: Have you ever tried cooking an egg on your HSF? Tasty. :tongue:
 
Crashman

Crashman

Polypheme
Former Staff
Dec 31, 2007
48,594
91
66,190
Look for my post on the subject. What your trying to accomplish is a heat pipe. I gave a lot of information as to how a heatpipe could be turned sideways to use a radiator.

What's the frequency, Kenneth?
 
D

dejay

Distinguished
Oct 18, 2001
36
0
18,530
I had an idea of trying mercury, but I have no real idea if it would be effective. I do know that it would be expensive and extremely dangerous, but the idea of liquid metal running around my computer and taking it's heat away sounds pretty cool (pardon my pun).

<font color=purple>I'm new here, I don't deserve a sig.</font color=purple>
 
D

dejay

Distinguished
Oct 18, 2001
36
0
18,530
Ha! Too late, I'm already insane (and I'd be surprised if my family weren't).

<font color=purple>I'm new here, I don't deserve a sig.</font color=purple>
 
H

HonestJhon

Distinguished
Apr 29, 2001
2,334
0
19,780
not that kinda crazy....were talking serious crazy....
heavy metal poisoning is not cool.....
read a national geographic at my grandmas house once...the pictures were pretty bad...it is kinda sick.
but...feel free to try..hehe...let us know what happens....if you do...but then again...kinda expensive, and i think you would have to fill out some crazy ammounts of forms to get enough mercury to cool your cpu.....

-DAvid

-Live, Learn, then build your own computer!-
 
M

Mnx4

Distinguished
Aug 10, 2001
324
0
18,780
Actually mercury would be a bad choice because it has a lower specific heat than water and would not be able to take in much of the heat produced by the processor. You need to find something that has a higher specific heat and a little less poisonous. Why not antifreeze? heh heh.

Do you work at UPS? 'cuz i could have sworn you were checkin out my package!
 
D

digital_trucker

Distinguished
Nov 7, 2001
559
0
18,980
And apparently it doesn't conduct heat as well as straight water--at least, so I've been told.

Wabbit season! Duck season! AMD season! INTEL season!
Hmmmph--whatever...
 
G

Guest

Guest
Actually antifreeze is used in computer systems to help prevent corrosion (just like it does in cars), especially in cooling systems with both copper and aluminum where there can be a battery effect. Antifreeze is also used in cooling systems to help prevent things from growing in cooling systems. Waterwetters are also good for these reasons too. Antifreeze does conduct heat less than just plain water, distilled water always works best. A favorite "recipe" is that of overclock-watercool's found <A HREF="http://www.overclockwatercool.com/coolant.html" target="_new">here</A>. I've seen the idea of rubbing alcohol being used, but the conclusion has always been that is wouldn't work as well. I've also seen some wacky stuff like submerging the whole PC in non-conductive liquid and using dry ice and liquid nitrogen, but that is a good way to ruin your PC.
 
P

peteb

Distinguished
Feb 14, 2001
2,584
0
20,780
Water wetter is fine, but true antifreeze is corrosive and should be flushed out of car systems fairly frequently. It does prevent growth of nasty stuff, but so can other, non-corrosive substances. This is why generally car coolant systems should be drained in summer and replaced with a non-antifreeze solution generally. There are new products out now that do the same thing whilst being much less corrosive (I'm not talking sulphuric acid in any case) and they claim you do not need to flush them...

-* <font color=red> Under Offer </font color=red> *-
email for application details
 
E

Ed_Phoon

Distinguished
Dec 31, 2007
275
0
18,780
Look at <A HREF="http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1297886772" target="_new">this</A>.
I'm sure the 3M Fluorinert is better than water.
 
H

HonestJhon

Distinguished
Apr 29, 2001
2,334
0
19,780
yeah, but flaurinert is EXPENSIVE!
but flaurinert can be cooled with ln2 i think, and then used to cool the system.
also, you will have to replenish the ln2...
and unfortunately, unless you are generating enough heat to heat up the flaurinert after it is cooled down, i am pretty sure that it turns into a gel like substance when it gets too cold...
that would not be very good for the pump, or the cooling capabilities of the total system.
not sure tho.

-DAvid

-Live, Learn, then build your own computer!-
 
M

Mnx4

Distinguished
Aug 10, 2001
324
0
18,780
yeah i have to agree that fluoronet is really cool. Id love to get me some of that!.. if i had the money. But off the subject of liquid cooling, why not solid cooling? i know it sounds weird but why not use a really cold substance? Theoretically, couldn't one put a small block of Dry ice (solid CO2) on the proccesor or in front of a fast fan? I don't think it will form a condensate because it sublimes so it would have no time to become a liquid. Either way i dont think a CO2 would be conductive, but i actually don't know. Any ideas, comments, questions?

Putting the laughter back in to Slaughter.
 
H

HonestJhon

Distinguished
Apr 29, 2001
2,334
0
19,780
that would be a very high maintenance system.
imagine replacing the dry ice like once a day...
but it might cool the air good..
and cool air would make the heatsink cool better.
what i am thinking about is this.
a refrigeration system, that cools the air to a very low temp, but right before the air goes into the system, there is a dehumidifier, which dry's the air so that it will not be as likely to condense.
i dont know how good that would work, but it might cool the system down a bunch!
anyone else have an idea about this?

-DAvid

-Live, Learn, then build your own computer!-
 
K

knowan

Distinguished
Aug 20, 2001
991
0
18,980
Your setup wouldn't work. Condensation isn't caused by cold humid air hitting a hot surface, it is caused by warm humid air hitting a cold surface. Running a defumidifier on the cold air is pointless, you would have to de-humidify the warm (room temp or ambient) air instead.

In theory if you had an air conditioner and a pipe, you could just pipe the cold air into the computer and not worry about condensation. So long as the air conditioner blows in the cold air then the warm moist air cannot enter. The only condensation you would get would be on the outside of the case since the only only air that could enter the case would be coming from the air conditioner. Of course you would either have to have the air conditioner running constantly or else turn it on about 10-20 min before you want to use the computer to allow it to cool down/dry out (interesting, you need time to let it "warm up" to cool down).

--------------
Knowan likes you. Knowan is your friend. Knowan thinks you're great.
 
M

Mnx4

Distinguished
Aug 10, 2001
324
0
18,780
yeah i do realize that it would be a high-upkeep system, but of course i was just posing the theory. It would cost a bunch to get a new block of dry ice each day. As for the problem of condensation, couldn't you keep the cold air in a metal pipe with sealed ends and just drop a chunk of dry ice in there every once in a while? There may be however a condensate that will form on the pipe itself. So i guess there really isnt a point to go through all the trouble of get a dehumidifier, more dry ice, etc.

Putting the laughter back in to Slaughter.
 
G

Guest

Guest
Ideas? Just that the rapid cooling of the dry ice touching the chip would probably crack your CPU and maybe parts of your motherboard too. This stuff has been done before: submerging the PC in flouranert and setting blocks of dry ice in it then circulating liquid nitrogen through it. I forgot the link to it, but it was what I was referring to in my last post. Amazing enough it all worked quite well until the liquid nitrogen was used. Here is a link to another time it was done where they skipped the dry ice and had little problems:
<A HREF="http://www.octools.com/index.cgi?caller=articles/submersion/submersion.html" target="_new">http://www.octools.com/index.cgi?caller=articles/submersion/submersion.html</A>
 
B

bum_jcrules

Distinguished
May 12, 2001
2,186
0
19,780
I just read that entire article. That is some crazy stuff. There has to be an easier way.

<b>All for one and one for all...and 3 for 5! - Curly - The Three Stooges</b> :lol:
 
silverpig

silverpig

Splendid
Dec 31, 2007
5,068
0
25,780
Water is just about the best coolant you can get. It has a high specific heat capacity, and it conducts well. It is even the best choice for a vapour/phase change coolant system as it has a very high latent heat of condensation/evaporation.

The only thing that would be better than straight water is to add some water wetter. Anything else would have more negative effects than positive.

A mercury coolant system would also be VERY heavy.

Lyrics. Wasted time between solos.
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom