1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Idea/project chilled water cooling FITS

Discussion in 'Overclocking & Cooling' started by Fitseries3, Jun 15, 2008.

1. intel igentNew Member

Joined:
Jun 5, 2005
Messages:
4,641 (1.25/day)
434
Location:
you would think so....

care to elaborate on how exactly you attempt to do this?

do you just plan to place a coil of copper tubing in a bucket, fill with water and add DICE?

2. imperialreignNew Member

Joined:
Jul 19, 2007
Messages:
7,043 (2.40/day)
909
Location:
Sector ZZ₉ Plural Z Alpha

You're not thinking of Pascal's law, but I can't remember the name of the physics law you're thinking of . . .

Pascal's law is the basis of hydaulic theories and practice; pretty much - if fluid exists in an enclosed system, and pressure is applied to the fluid at one end of the system, the fluid will exert an equal or great amount of pressure amoung all points of the system. A fluid cannot be compressed.

I think you're thinking of if fluid is in an enclosed system and temperature of the fluid changes, so will it affect the pressure of the fluid within the system due to thermal expansion and contraction - but this shouldn't be anything to worry about if there is still ample amount of space left in a reservoir, as it would allow room for the fluid to expand and contract.

Or, you might be thinking of Newtons' law of cooling, which states "the rate of heat loss of a body is proportional to the difference in temperatures between the body and its surroundings, or environment." and is the basis of convection cooling theory - which means that as fluid accepts heat, it will flow to a cooler area where the fluid is then cooled, and cool fluid will be drawn to areas of heat. This is general convection theory. Warm is drawn to cold, as cold is drawn to warm.

Or, you might be thinking of Fourier's law, which states "that the time rate of heat transfer through a material is proportional to the negative gradient in the temperature and to the area at right angles, to that gradient, through which the heat is flowing" - which is the basis of heat conduction theory, which pretty much means that the amount of heat that can be transfered is dependant upon how well the medium (fluid) can transfer heat, and how quickly that heat can be given back off as well.

I think what you've been trying to describe is pretty much . . . the cooling system can only be as efficient as the fluid you're using in the system; I might be wrong, though.

But, don't forget that Murphy's Law supersedes and surpases all other laws, even Newton's law of gravity and Einstein's theory of relativity.

3. moocow0463

Joined:
Dec 28, 2007
Messages:
421 (0.15/day)
14
i also happen to be a huge gear head so heres my .02 cents, dont use antifreeze water has a better heat absorbtion rate then antifreeze, example: people who live in arizona run pure water, not pure antifreeze, if you want colder temps go to a performance shop or maybe even checkers or autozone would carry it, ask for "Water Wetter" in a cars radiator mixed with water the temperatures can drop up to 10*F which is pretty substantial when we are talking computer parts

4. imperialreignNew Member

Joined:
Jul 19, 2007
Messages:
7,043 (2.40/day)
909
Location:
Sector ZZ₉ Plural Z Alpha
I would tend to agree - I actually forgot to mention the antifreeze bit earlier, but I hinted at it - but, IMO, water wetter performs it's best when the system is pressurized and when it has to deal with the insane heat we are used to with cars. It'll still provide some advantages, but I don't really think you'd see the same benefits as you would with an automotive cooling system.

My personal opinion, would be to use distilled water, as it has better thermal conductivity than standard tap water does. There are also a ton of water additives for computer cooling systems, but you'd need to do the research on these, as some are tripe, some don't do squat, and some are extremelly overpriced for just food colouring. But there are some really good ones out there, though.

5. moocow0463

Joined:
Dec 28, 2007
Messages:
421 (0.15/day)
14
ya true water wetter may not work as good in a low to zero pressure system but i think it would be worth testing its not very expensive and i would be intrested to see what it does

6. driver66

Joined:
Jun 4, 2007
Messages:
1,053 (0.35/day)
115
Location:
indiana
The problem with that is if he does it with DICE he may see below freezing temps that even with the water flowing may slowly freeze. That would be a big problem lol

Fitseries3 says thanks.
7. imperialreignNew Member

Joined:
Jul 19, 2007
Messages:
7,043 (2.40/day)
909
Location:
Sector ZZ₉ Plural Z Alpha
I would also tend to agree - but only in a static system. If there is actually a pump helping move fluid along, I don't forsee any freezing action - unless there is a pocket of fluid in the radiator that isn't moved, but incoming warm fluid should still prevent ice formulation.

TBH, I don't think ice would even have a chance to start forming until the whole system temperature has equalized to below 0C. Like in an automotive system, if the vehicle has been shut down straight water in a motor won't start to ice until the whole system's (radiator, motor, heater core, hoses, etc) ambient temperature is at or below freezing.

8. intel igentNew Member

Joined:
Jun 5, 2005
Messages:
4,641 (1.25/day)
434
Location:

i would think for this to be effective you would need to have a very large coil of flat tubing, in a very large reservoir (insulated metal) filled with acetone or similar (catalyst for the DICE) and a strong pump to force the liquid (some form of alcohol) through the tubing and a large amount of DICE on hand?

and of course the line's would have to be insulated.

hmmmmmmm........

File size:
54.2 KB
Views:
74
9. SkitzoNew Member

Joined:
Apr 3, 2008
Messages:
720 (0.27/day)
86
in another thread someone suggested mineral oil for sub zero cooling. It was more to address condenstaion I think though. I guess it would depend on how it reacts to the low temps, it may need to be thinned but is there something to do that? Would alcohol work. It evaporates at 30.5*C but that shouldn't be a problem in a pressurized loop at subzero temps. I think the biggest "if" would be the pump, can it handle the extra work? I can't say how thermal transfer would work but it shouldn't freeze. Any thoughts?

EDIT: If the case is sealed and a second rad wad added inside the case for purpose cooling recirculated air, it should help with condensation regardless of the cooling fluid. Putting it first in the loop would prolly be best too.

Last edited: Jun 15, 2008
10. Fitseries3Eleet Hardware Junkie

Joined:
Oct 6, 2007
Messages:
15,509 (5.44/day)
3,107
Location:
Republic of Texas
as much as this stuff is gonna cost i think i may have a change in plans here.

im gonna just try ice in a bucket with the rad to see how well it works.

eventually.... when i get this AC rebuilt i'll have a true water chiller. but i havent decided whether to build a chiller or a straight up phase unit yet. either way... condensation is a bitch.

11. ZenEffectNew Member

Joined:
Mar 12, 2008
Messages:
622 (0.23/day)
86
dry ice will turn antifreeze into goop. use denatured alchohol in your loop or you will get freezing problems. also you have to worry about condensation on the tubing.
if you want some REAL fun.

Last edited: Jun 15, 2008
12. intel igentNew Member

Joined:
Jun 5, 2005
Messages:
4,641 (1.25/day)
434
Location:
mineral oil would be good for an immersion setup as that's what i recall seeing.

DICE placed in water create's a fog machine like effect? correct or not? but it would not be as good a thermal conductor than acetone in this situation?

this is a good topic where is everyone?

fit's don't give up the idea

DICE is pretty cheap and so is acetone, you could prolly build something like this for pretty cheap?

13. ZenEffectNew Member

Joined:
Mar 12, 2008
Messages:
622 (0.23/day)
86
honestly w/ dice i think its better to just make a dice pot out of parts. intel stock cooler copper base + some copper/aluminum tubing + some solder = dice pot.

14. SkitzoNew Member

Joined:
Apr 3, 2008
Messages:
720 (0.27/day)
86
damn, i knew that. happened in my z-28 one winter lol, last time I let a friend change my coolant...

15. ZenEffectNew Member

Joined:
Mar 12, 2008
Messages:
622 (0.23/day)
86
he probably added too much water.

16. Fitseries3Eleet Hardware Junkie

Joined:
Oct 6, 2007
Messages:
15,509 (5.44/day)
3,107
Location:
Republic of Texas
ok... here's the idea...

take an old window AC and carefully take it apart. you'll find the 3 main components inside...
1. pump/compressor
2. condenser (hot side)
3. evaporator (cold side)

you can use the AC's original evaporator but depending on what you are doing it may be better to build your own(like in the pic below)

basically, you submurse the evaporator in a bucket, cooler, whatever you have, full of the coolant for the water loop. the evaporator supercools the water.

17. ZenEffectNew Member

Joined:
Mar 12, 2008
Messages:
622 (0.23/day)
86
YES, MY THOUGHTS EXACTLY!

18. Fitseries3Eleet Hardware Junkie

Joined:
Oct 6, 2007
Messages:
15,509 (5.44/day)
3,107
Location:
Republic of Texas
the upside to this method over a regular phase change is that you can cool your whole loop and not just a single part.

19. intel igentNew Member

Joined:
Jun 5, 2005
Messages:
4,641 (1.25/day)
434
Location:
i agree, but that was not the original idea......

funny; now it seem's as though the real question's are being avoided and the "plan" is to go to a standard chiller, which of course is more feasible and prolly more efficient, but never the less.....

20. SkitzoNew Member

Joined:
Apr 3, 2008
Messages:
720 (0.27/day)
86
I've yet to take apart an ac unit. so i have a couple questions about this idea. The bucket would in a sense be your res and the rad in the picture is for the ac parts not the pc loop. This negates the need for a rad in your pc loop. is this correct? What was the btu rating for the ac?

yeah your right intel igent, we can figure a way to make the original idea work.

21. Fitseries3Eleet Hardware Junkie

Joined:
Oct 6, 2007
Messages:
15,509 (5.44/day)
3,107
Location:
Republic of Texas
well.. for the sake of the threads original top i still do plan to try the rad in a few different cooling liquids. DICE is cheap as hell around here and the practically give it away. i have a 400l tank of LN2 sitting here too that i need to get around to playing with.

the AC cooler will come later(very soon) but i was just expanding on the idea that ppl were getting at. i have had this planned for a while now.

the whole purpose of this thread is to show ppl that a total crap setup can be used to cool your computer down to sub ambient temps to allow for better OC's for benching.

22. Fitseries3Eleet Hardware Junkie

Joined:
Oct 6, 2007
Messages:
15,509 (5.44/day)
3,107
Location:
Republic of Texas
the ac portion it separate form the computer's loop. the frosted coil in the pic cools your water in the "bucket" and there is no need for a rad in the loop.

23. SkitzoNew Member

Joined:
Apr 3, 2008
Messages:
720 (0.27/day)
86
the water wetter that was mentioned, I've used a product by that name working at a tree nursery. It was explained to me that it was basically liquid soap with a few ingredients missing. is this the same product?

24. Fitseries3Eleet Hardware Junkie

Joined:
Oct 6, 2007
Messages:
15,509 (5.44/day)
3,107
Location:
Republic of Texas
wetter water just lowers the surface tension of water. i think it uses ethylene glycol.

anyone know where to buy a tiny compressor? one smaller than a AC has.

25. intel igentNew Member

Joined:
Jun 5, 2005
Messages:
4,641 (1.25/day)
434
Location:
i'm reffering to these comment's within regard's to making a "simple" chiller

you said it would be easy? we started discussing it and now youv'e kinda changed your mind?

or deviated from it?

see where i'm coming from?

physics is a B1TCH! but definately interesting.....

let's keep on the topic of thermal dynamic's, we can only benefit

EDIT : watter wetter is useless in PC cooling