Why does watercooling produce less heat? [Impossible.]

The alternative to direct air cooling

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Why does watercooling produce less heat? [Impossible.]

Post by MoJo » Fri Aug 24, 2007 11:43 pm

I have been trying to get my head around this one for ages. My system is an Athlon X2 4400+ and GeForce 7800GT. Initially they were on air cooling, now on water. When I moved to water, the amount of heat the system was throwing out into the air dropped dramatically.

The whole system runs cooler. The air pushed out of the case is cooler. My room doesn't get as hot. Yet, the amount of heat energy must be constant. :?

[Mod: your last statement is correct: by definition it MUST put out the same heat; no matter how it is cooled.]

AZBrandon
Friend of SPCR
Posts: 867
Joined: Sun Mar 21, 2004 5:47 pm
Location: Phoenix, AZ

Post by AZBrandon » Sun Aug 26, 2007 11:04 am

Aren't all watercooling radiators outside the case? I've never seen one that's inside the case, hence the water took all that heat from inside and moved it outside the case, so there's less heat inside the case to be blown out per minute.

nutball
*Lifetime Patron*
Posts: 1304
Joined: Thu Apr 10, 2003 7:16 am
Location: en.gb.uk

Re: Why does watercooling produce less heat?

Post by nutball » Sun Aug 26, 2007 11:39 am

MoJo wrote:When I moved to water, the amount of heat the system was throwing out into the air dropped dramatically.
This is an incorrect statement, especially when combined with your other assertions. Thermodynamically you need to regard the "system" as PC + cooling system together.
The whole system runs cooler. The air pushed out of the case is cooler.
This is quite possible depending on your setup (eg. if you're using a Reserator). That's entirely different from stating that the entire system put out less heat.
My room doesn't get as hot.
Again, either incorrect or anecdotal. Especially given the weather here recently.
Yet, the amount of heat energy must be constant. :?
It is. Hence my statements above.

~El~Jefe~
Friend of SPCR
Posts: 2887
Joined: Mon Feb 28, 2005 4:21 pm
Location: New York City zzzz
Contact:

Post by ~El~Jefe~ » Sun Aug 26, 2007 11:59 am

Well, I mean technically the less heat that is generated in the case, the less electricity it uses.

I dont think this applies however to your system in a noticable way though.

PSU's exhibit this the worst/noticably. More heat evolved, the efficiency drops like a stone. That is why 1000 watt kiddie shit psu's make no sense as the more you use them, the hotter they get, the less wattage they put out, the more they use, so you basically are left with a room heater.

however, I dont think this could apply to a cpu and gfx card. Maybe I am wrong, I dont know :)

I do know that my Res1 blue boy tower heats my room up considerably on my "gaming rig" pc that is on 24/7

I just can't figure how it would make less heat. Hm. Well, when you pull off the cpu and gfx card heat from the case, the PSU stays very cold running. that would give you like 20 watts benefit?

MoJo
Posts: 773
Joined: Mon Jan 13, 2003 9:20 am
Location: UK

Post by MoJo » Sun Aug 26, 2007 6:53 pm

Thanks for the comments, they are pretty much what I was thinking.

My radiator is inside the case (it is a server case, basically like two normal cases stuck side by side with the mobo in the left hand chamber and the PSU/HDDs/radiator in the right chamber.)

~El~Jefe~, your comments about the PSU are interesting. It certainly runs cooler thanks to not sucking hot air off the mobo/CPU. But as you say, it could only be a small difference, probably offset by the heat generated by the water pump.

I think the expelled air being cooler could be down to heat being spread more evenly over a larger volume of air. Before, heat from the GPU and CPU were being blow out largely by one fan, so that air was very hot. Now the heat is distributed a bit more, for example by the radiator which has a much larger surface area than the old HSF.

In the end though, it must be the same volume of heat energy, as nutball says. Thing is, with the same outside temperature, doors and windows closed, central heating turned off and no other obvious variables, my room is about 3C cooler after 8 hours at night. I wasn't in the room during that time either, so it wasn't my body heat. Thermometer was wall mounted away from the computer.

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Mon Aug 27, 2007 1:09 am

I believe MOSFETs etc are slightly more efficient when they are kept cooler, so it is possible that the computer is actually producing less heat.

~El~Jefe~
Friend of SPCR
Posts: 2887
Joined: Mon Feb 28, 2005 4:21 pm
Location: New York City zzzz
Contact:

Post by ~El~Jefe~ » Tue Aug 28, 2007 6:28 pm

hey, that is true as well.

theres actually a lot of places to kill electrical efficiency. watercooling is the best to cure it I feel.

Aris
Posts: 2299
Joined: Mon Dec 15, 2003 10:29 am
Location: Bellevue, Nebraska
Contact:

Post by Aris » Tue Aug 28, 2007 9:11 pm

more than likely what you are experiencing is the fact that water is a more effecient means to transfer heat than air. So it moves the heat from the system components, to the radiator and out of the case quicker and more effeciently than it would with air. This is why the air comming out of your system "feels cooler". It didnt spend as much time inside the system heating up to a higher temp.

As far as your room not "feeling as hot", again its probably cause you have more dispersion of the heat in a quicker amount of time, so the heat isnt so isolated in and around the system and is merged with the ambient temps more casually. If that makes sense. Im not really sure how to say it.

The overall heat output of the system though is still the same.

Post Reply