Cool looking water cooler!

[Acethechosenone]

The same energy required to heat the larger mass is now needed to cool that large mass back to normal.

You are absolutely right, I overlooked the fact that if a larger mass takes longer to heat up similarly it will take longer to cool down, therefore the size of the reservoir will only increase heating and cooling times.Thanks for pointing that out.

 

[tool_462]

I was meaning to imply that the difference isn't enough to affect the performance. When I say it takes 40 seconds to cool down, I idle at 17C and load around 27-29 depending on load. It drops within a couple seconds to ~21, ten more seconds to ~19 and after roughly 40 seconds it gets back to 17.

 

[rubix_1011]

I know its not the best; I had a Chevette heatercore on an older system that ran great. I found that the 2x120mm I have fits the back of my case perfectly, uses the casefans I already had, and I don't have to mod/cut/drill to get a new rad to fit. For my needs, it works very well. My biggest concerns were to keep my CPU down in temps (the 640's were hot, the 641's fixed that) and my GPU is notoriously hot (as are most these days). At stock, the P4 was running at 38C idle, 52C under full load/games and the 7800gtx ran 42C idle, 62C load. Neither sees over 40C or 38C. Consistency and cool are what I wanted.

Thanks for the link though...maybe I can find a 2x120 that is similar.

 

[SuperFly03]

Go w/ the new OCZ phase changer, only 300 bucks

I'd love to but where the hell is it? Been waiting for nearly 2 years (*i think*). I am all about phase change, but honestly I can't afford it right now. I can afford $300, but $800 just for cooling is a bit out of my price range (stupid new car savings lol).

 

[clue69less]

finally someone that has more than half a brain

Watch your mouth. You've written stuff in this thread that ranges from just plain wrong to sloppy, so don't be bragging about your brain. Oh, wait, 51% is more than half a brain, so there you are.

 

[tool_462]

:?

 

[rubix_1011]

Wow...this is really becoming a crap-shoot.

 

[rubix_1011]

I'm not offended...its just becoming a pointless pissing contest.

 

[rethdog]

I understand the principle of it, I used to have car with sodium in the exhaust valves, under heavy use the sodium liquidfied and conducted the heat energy more effectively to the edge of the valves than a solid valve ever could. Same principle here but the temperature differential is so small in this application (CPU to ambient) it's hardly worth the effort, just fit a bigger slower rotating fan with nice copper / aluminum heat sink. Remember the fan can spin too fast, as the fins on car a radiator are crimped to slow the airflow down.

 

[clue69less]

alright, I'm sorry, didn't think it would offend you that much, I was just really getting anoyed at us not getting to understanding each other, either way, one of has to be wrong, as tool agrees with me, but what you say is that I'm wrong, I think someone has to be the judge of this to see who is finally right

OK, no problem. The thing is, you're saying different things now than when I got into this little argument with you. I pretty much took your words as they were written when you replied to Granite. You can try to call in all the judges you want and it won't change my mind. Go out and do what I described - try to build an effective small coolant volume loop. Have you done any water cooling at all?

 

[clue69less]

Same principle here but the temperature differential is so small in this application (CPU to ambient) it's hardly worth the effort, just fit a bigger slower rotating fan with nice copper / aluminum heat sink.

I disagree. A good quality WC loop will cool better than any commercially available HSF that I've ever seen or read about. Part of the reason that good WC systems do so well is the surface area of the rad in contact with cool air. It's not uncommon for people to use rads the size of three 12cm fans. Heat pipes are pretty amazing to me but I'm not convinced that having equal airflow would get a HSF system to compete well with WC because of the differences in the heat transfer process between a HSF/heat pipe block and a WC block. The WC block heat transfer mostly involves conduction and convection from the block into the moving fluid. In a heat pipe, vaporization, condensation and surface-tension-driven transport are key processes in the heat transfer. I've never seen a good quantitative treatment for the range of heat from a CPU IHS surface, but my gut instinct is that WC has better potential for heat removal. Loads of satisfied overclockers seem to agree.

 

[rethdog]

taco is right, the volume of water has no effect on the overall efficiency of the movement of heat energy. (other than a larger reservior acts as radiator).

but the increased volume of water could act a buffer, so if your processor had speed step control and it was cycling between low and high heat the larger volume of water would delay the onset of a stead state temperature, or reduce it.

so you are both correct if this makes sense?

 

[clue69less]

not my own, but I have done it with other peoples coolers for them (n00bs these days, don't know how to connect the tubing to the water blocks) and from what I have tried as long as the the water amount is a considerable level and filled up, the performance stays the same with the exception if you change the rad or pump

Sure, but that's a long way off from what you were saying a page ago:

"The amount of liquid has a very small fraction to do with how well it cools."

So to test your comment above, go design a WC loop that uses 1/10th or 1/50th as much fluid as a typical 2x120 rad. If the amount of liquid has a very small fraction to do with how it cools, then just go build a small volume system and test your hypothesis! What you'll find is that as you drop the liquid volume, the size of the rad has to decrease. All along, I've been making the point that for any required amount of heat removal, a minimum rad surface area will be required depending on the ambient temp and max allowable CPU temp. That minimum rad size will dictate a minimum liquid volume and as you miniaturize, problems with pumping and sealing increase.

 

[rethdog]

you disagree with what? read my post i didn't say a WC loop was no better, I made the point it wasn't worth the effort for the average Joe, we weren't talking over clocking.

yes bigger rads will help you so far, but you ultimately become limited to your point of contact on the CPU.

so yes a liquid near the heat transfer point is better, that's my point with the sodium in the valves

 

[clue69less]

you disagree with what?

I quoted exactly what I disagreed with:

"Same principle here but the temperature differential is so small in this application (CPU to ambient) it's hardly worth the effort, just fit a bigger slower rotating fan with nice copper / aluminum heat sink."

What I'm getting at is that if you try to push an OC on air, you'll end up at a lower VCore and OC than you will with a good WC system.

read my post i didn't say a WC loop was no better, I made the point it wasn't worth the effort for the average Joe, we weren't talking over clocking.

But we weren't not talking overclocking either.

yes bigger rads will help you so far, but you ultimately become limited to your point of contact on the CPU.

Yep. I expect that eventually, CPUs will incorporate an improved cooling interface for that very reason.

 

[clue69less]

by incorporate a better cooling surface, what do you mean, like better contact to the core, or a better heatspreader?

It could go a bunch of different ways. It's possible to make a chip with internal fluid-filled or Cu-filled channels. One could incorporate a three-dimensional IHS, etc. I've seen mil-spec power transistors with good heat transfer built in, so why not CPUs? Ah- it's probably all just wishful thinking. With heat loads decreasing on mainstream CPUs, it will probably be less and less of an issue in the future.

 

[clue69less]

that would be nice, because I'm tired of having to pay extreme prices on some cooling options without getting enough performance increase to justify it (extreme to my budget anyway)

Your link has given my kids a new anthem.

 

[clue69less]

which link?

The one at the bottom of your sig. You tube FtS tune. FtS, FtFSS.

 

[berniemansell]

Hi guys Im looking to set up a water cooling system for my AMD6000+ I have allready purchased a Thermaltake AquaBay M3 1U Liquid/Water Tank, and am now looking for a Radiator. I wish for a good one because i will be upgrading my system to the AMD Bulldozer when it is released. i would like to use my watecooling for my new computer,
With my current system i will be cooling a AMD 6000+ CPU & Nvidia Geforce 9600 512mb, The Graphics Card is overclocked, but because i'm cheap i have a MOBO that doesn't allow me to overclock my cpu.
Are there any reviews on this radiator, http://www.koolance.com/water-cooling/product_info.php?product_id=634

Or does any one know of any other good radiators

Thankyou

Bernie

 

[Conumdrum]

LOL, why ask on this thread, did you read the last post date, this is a crazy old thread. Anyway, on to your question...........

Koolance rads are very poor and shunned in the WC world.


First you gotta learn about WC. It's not like walking into Best Buy.
Spend a while (weeks is best for your sanity) at these links.
Look at the hundreds of loops close to your case and components in the stickies, read a couple 50 or so threads over the next week or so, you'll be on the ball to make the right choices and by then know how to put it together.
Not 'Roket Sience', but basic knowledge is required.
And you should spend a few hours on this site reading threads. It's how we learn. Once the goodies show up on your doorstep your on your own.
For your benefit please spend a few days reading a LOT. At the busiest places for WC masters. Guys who have done it for YEARS at OC forums and xtreme forums. It took me a while (I was OCing on air, aftermarket stuff, bios settings, best chipsets etc etc) to learn the language and the tricks to a easy install.

Don't expect miracles or SUPER DOOPER over clocks. What you will get is a quiet system that can handle OC to the max of your hardware IF you buy quality and buy smart. And minor maintenance too, a bonus for the water cooler.

Also while there please read on case mods etc. The radiators are not for small cases, pumps and hose routing, wire management and other things are important. Google your planned case and the word water-cooled in one line. You might get lucky.
.............................

Edit: The next paragraph was from 2008. With the advent of the HOT i7 and bigger GPU's, it has changed. A 200 size MIN for an i7, you want big overclocks, better go 320 sized rad.
...................................

IF you just cool your CPU and your NB if you want, you can get by with a 120.2 sized radiator (RAD). And MAYBE fit in inside depending on your mod skillz. You want to cool your GPU too, you'll need a 120.3 sized rad, and it probably won't fit inside. The rear external rad really works great. No matter what your adding 15lbs to your PC, maybe 20lbs.

Once you got an idea of what is good/bad then start getting your system for WC put together and we'll be glad to help.

http://www.xtremesystems.org/forums/index.php? Not a noob site, but great stickies
http://www.ocforums.com/ My fav, good peeps, know their stuff, less hardcore
http://www.skinneelabs.com/MartinsLiquidLab/
http://www.over-clock.com/ivb/inde [...] opic=20277 A GREAT Europe site
http://www.overclock.net/water-cooling/
http://translate.google.com/transl [...] n&ie=UTF-8 Info on rad testing
http://skinneelabs.com/
http://forums.extremeoverclocking.com/showthread.php?t=282232

Stores
http://www.dangerden.com/index.php [...] e&Itemid=1
http://www.petrastechshop.com/
http://www.sidewindercomputers.com/
http://www.jab-tech.com/