First off I'd like to apologize if this has been discussed. I've been searching/reading the forums for the past few days and I haven't seen this discussed.
As you can probably guess from the fact that I’m even on this forum I am trying to get my computer as silent as possible without sacrificing performance. I am running an overclocked AMD 4200 x2 and an overclocked ATI 1900XT. I was looking into passive cooling but I’d rather use the radiator I have (active) if I can…
We all know that water takes away heat much better than air which is why water cooling is used. When we watercool we use a radiator to take the heat from the water then use air to take the heat from the radiator. Since water does a much better job dissipating the heat why not use it to take heat away from the radiator?
There has to be a reason not to do this because no one is doing it from what I can tell.
However I am in the middle of testing right now. With silent 600 RPM fans on my radiator my CPU would quickly get up to 57 degrees Celsius in about 30 minutes while running both cores at 100%. With loud 2000 RPM fans the temp would stabilize at 49-51 degrees after the same amount of time. Right now my radiator is fanless and sitting in a tub full of water. Not ice water or anything just tap water. After 30 minutes of testing it reached 44 degrees but got there very slowly.
A couple points:
I know my test time is inconclusive but I interrupted it because I wanted to get this discussion started. I will continue testing as soon as I post this.
I know that the water will get saturated with heat eventually but how often is the CPU running full load for extended periods in real world application. I know for myself it’s almost never. Also if the water container was aluminum then it would dissipate heat from the water it’s holding I assume.
Like I said this sounds too easy so there must be a downside. Please let me know what it is. If you have already tested this idea please also let me know your results.
Thanks, I’ve really enjoyed this forum so far. I didn’t know there were this many people like me that were so into silencing!
*edited for grammar*
Am I missing something? (new idea on passive cooling)
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
unimatrix0
- Posts: 88
- Joined: Sat Jul 15, 2006 2:07 pm
Hey welcome 
I think you've lead yourself into a passive solution without necessarily looking for it. An aluminum container would dissipate heat so it’d be passive cooling. I bet that if you attach aluminum heat sinks all around the container it’ll enhance the cooling. The caveat is that such a container isn’t amazingly practical. Do you place it internally or externally? And how will you deal with evaporation? An open container will evaporate water like crazy.
Take it a step further. Make a closed container, attach two fittings, and put heat sinks around it. You’d now have a sealed passive radiator. Now that’s unexplored territory. There are some passive solutions such as Reserator and Innovatek’s, but I think there is a lot more to be exploited on the subject.
My take is to make a large thin waterblock panel, and attach a giant heatsink panel on top. It’d be mounted on the side of a PC. So water would have a giant area to go through. So far I haven’t been able to find the heatsink panels. But the quest continues…
I think you've lead yourself into a passive solution without necessarily looking for it
. An aluminum container would dissipate heat so it’d be passive cooling. I bet that if you attach aluminum heat sinks all around the container it’ll enhance the cooling. The caveat is that such a container isn’t amazingly practical. Do you place it internally or externally? And how will you deal with evaporation? An open container will evaporate water like crazy.Take it a step further. Make a closed container, attach two fittings, and put heat sinks around it. You’d now have a sealed passive radiator. Now that’s unexplored territory. There are some passive solutions such as Reserator and Innovatek’s, but I think there is a lot more to be exploited on the subject.
My take is to make a large thin waterblock panel, and attach a giant heatsink panel on top. It’d be mounted on the side of a PC. So water would have a giant area to go through. So far I haven’t been able to find the heatsink panels. But the quest continues…
Well it's been 2.5 hours now and the water has went from cold tap water to 35 degrees Celsius. My CPU temp has very slowly crept up and is now at 55 degrees.
My plan is to get a sealed aluminum box and run the lines through same size rubber grommets. It will be external and my Antec 180 will not be moved. I don't know if this will be tight enough to eliminate or substantially reduce evap or not. I doubt it will evap faster than my old aquarius 2 and I got used to filling that thing up
My plan is to get a sealed aluminum box and run the lines through same size rubber grommets. It will be external and my Antec 180 will not be moved. I don't know if this will be tight enough to eliminate or substantially reduce evap or not. I doubt it will evap faster than my old aquarius 2 and I got used to filling that thing up
It's not so much because water dissipates heat better, but it has a much high specific heat capacity than air (ie it takes much more energy to raise the temp of a mass of water by 1 degree than the same mass of air).Since water does a much better job dissipating the heat
I can't see anything wrong in principle; ISTR reading about a guy who kept his radiator in a constantly running flow of cold water in a nearby stream! (this is bad for the environment however)Like I said this sounds too easy so there must be a downside.
-
datapappan
- Posts: 199
- Joined: Wed Nov 10, 2004 1:44 pm
- Location: Sweden
Great idea, you're buffering the heat in betwen CPU load sessions.
The downside with water is that things tend tp grow in it, so using distilled water and a hermetic cover would do the trick.
If you want to calculate how much water is needed, here's a formula:
1 litre of water will hold 4,182 Ws/C
This means that if you put in 300 W for 30 mins (1,800 sec) the water temperature will increase 300*1800/4182=129 degrees (!). However, the container will radiate heat, so the net load will be less than 300 W. On the other hand, as the water gets warmer, so will the cooling liquid, and as a result also the CPU.
Interesting to hear more from your experiments.
The downside with water is that things tend tp grow in it, so using distilled water and a hermetic cover would do the trick.
If you want to calculate how much water is needed, here's a formula:
1 litre of water will hold 4,182 Ws/C
This means that if you put in 300 W for 30 mins (1,800 sec) the water temperature will increase 300*1800/4182=129 degrees (!). However, the container will radiate heat, so the net load will be less than 300 W. On the other hand, as the water gets warmer, so will the cooling liquid, and as a result also the CPU.
Interesting to hear more from your experiments.
Thanks for the formula. I'll work with that before I look for a container...
After 3.5 hours the outside water reached 39 degrees and the CPU reached 59 degrees. Keep in mind the longest I ever ran both cores with the 600 RPM fans was about 40 minutes and the CPU temp reached 57 degrees. I don't know the ambient temp difference from that test as it was months ago but considering it the difference in CPU temp is negligible. Also with the way the CPU temp was rising in the test with the fans I have no doubt that after 3.5 hours the CPU would have been at least several degrees hotter.
After the 3.5 hours I let the computer idle overnight and now after 10 hours of idle the temp has settled at 39 degrees. I did not measure the outside water temp (forgot) but I will when I get home. That will give it 20 hours of running at running idle so the water should be as saturated as it will get with no load.
My most recent setup was 2 120mm fans running at ~1200 RPM and my idle temp was 36 degrees. Considering that my radiator is just sitting in a bucket of water and completely silent I think the 3 degree rise is pretty good. I will continue to play with it and post results. My new case, hoses, and hardware should be shipped today so hopefully I’ll have a water box setup within a week.
After 3.5 hours the outside water reached 39 degrees and the CPU reached 59 degrees. Keep in mind the longest I ever ran both cores with the 600 RPM fans was about 40 minutes and the CPU temp reached 57 degrees. I don't know the ambient temp difference from that test as it was months ago but considering it the difference in CPU temp is negligible. Also with the way the CPU temp was rising in the test with the fans I have no doubt that after 3.5 hours the CPU would have been at least several degrees hotter.
After the 3.5 hours I let the computer idle overnight and now after 10 hours of idle the temp has settled at 39 degrees. I did not measure the outside water temp (forgot) but I will when I get home. That will give it 20 hours of running at running idle so the water should be as saturated as it will get with no load.
My most recent setup was 2 120mm fans running at ~1200 RPM and my idle temp was 36 degrees. Considering that my radiator is just sitting in a bucket of water and completely silent I think the 3 degree rise is pretty good. I will continue to play with it and post results. My new case, hoses, and hardware should be shipped today so hopefully I’ll have a water box setup within a week.
I don't know if this is has been talked about here, but it's definitely not new. Closing the tub of water would stop the evaporative cooling effect. If you are going to close up the tub and make a metal box, you would be better off trying to make a proper radiator rather than just a giant water heatsink.
I don't know where it is now, but I'm sure a simple google search would find it, but this reminds me of the guy who used his fish tank for cooling, and as an added benefit it would warm the tank for his tropical fish.
I don't know where it is now, but I'm sure a simple google search would find it, but this reminds me of the guy who used his fish tank for cooling, and as an added benefit it would warm the tank for his tropical fish.