Radical CPU coolers from CoolIT

[bgerber]

With its Freezone and the Eliminator coolers, CoolIT brings hybrid Peltier/liquid cooling to the masses. How do the two coolers stand up against cooling systems using other methods?

 

[Gneisenau]

The main difference between the Freezone and Eliminator is the amount of cooling power. The Freezone is the more powerful of the two models, advertised to handle CPU's with a heat output up to 175 watts. The Eliminator is advertised to handle a heat output up to 175 watts.

Huh :?:

I assume only one is rated for 175 watts. Which one is it and what is the other one rated for? If I'm wrong, what makes one more powerful than the other one?

 

[mr_fnord]

Good article. It's interesting to see that plain old water cooling is actually better under load than Peltier-chilled water setups. Of course, part of this could be the prepackaged Peltier cooling unit vs the large, customizable water setup. Are there single-unit water setups like the Freezone and Eliminator, only without the Peltiers?

In other words, does the worse cooling performance of the Freezone come from the very limited size of the unit, or are the Peltiers just inefficient/not as effective at transferring heat as a plain old radiator? Fighting the heat inside of the case doesn't help I'm sure, but it would be interesting to see where the bottlenecks are, so to speak.

 

[gm0n3y]

While I admit I didn't read the whole article and you may have talked about this; there are three things missing from this article:

1) Overclocking. These kinds of coolers are pretty much solely used for overclocking. The real test of a cooler how good it is at higher thermal load.

2) Aftermarket air coolers. To me this is kind of an intermediate step between aftermarket air coolers and water cooling. I would like to have seen how this compared against a high-end air cooler (Turniq Tower, Scythe Ninja and the like). Even a high-end air cooler would cost less than 1/2 of these peltier coolers and it would make no sense to use one of these if an air cooler would provide similar results.

3) Noise. The only reason, other than overclocking, to use a system like this is because of noise. You should have included some graphs with noise measurements of all of the coolers at idle and load.

 

[warezme]

Oh-My-Gosh! 8O

What kind of "enthusiast" would only OC to 2.25 or 2.26 or whatever that lame number was, or use an E4300 to do it on?

Crank that sucker up over 3Ghz and tell us what voltages the CPU is running at than do some runs.

Its totally bogus. Its probably not built to handle more than a dull OC of 2.26Ghz and anything over that and the peltier will go chernobyl on you and fry your chip. Pffft, another lame review. 🙁

Stick 6800 on that overpriced thing and OC it and let me know what happens then.

 

[IcY18]

Oh-My-Gosh! 8O

What kind of "enthusiast" would only OC to 2.25 or 2.26 or whatever that lame number was, or use an E4300 to do it on?

Crank that sucker up over 3Ghz and tell us what voltages the CPU is running at than do some runs.

Its totally bogus. Its probably not built to handle more than a dull OC of 2.26Ghz and anything over that and the peltier will go chernobyl on you and fry your chip. Pffft, another lame review. 🙁

Stick 6800 on that overpriced thing and OC it and let me know what happens then.

Yeah really, I've seen too many articles like this testing the Freezone or Eliminator at such a mild OC it's so terribly frustrating. Who the heck would drop $300 just so they could overclock to 1/4 of the CPU's potential. Completely bogus is right, also while the author showed us a not-so-good Koolance water kit(compared to custom WC setup), he didn't show one high end air cooler?

This article fell about 30 yards short on a 40 yard field goal.

 

[cleeve]

1) Overclocking. These kinds of coolers are pretty much solely used for overclocking. The real test of a cooler how good it is at higher thermal load.

I did overclock the system 25%. Any higher and the stock cooler couldn't handle it at load, which would have made it impossible to bench against the stock cooler. I think there's value to that.

However, for you hardcore junkies I'll be using the Eliminator in an overclocking article coming out in a week where I'm pushing the e4300 as far as it will go.

2) Aftermarket air coolers. To me this is kind of an intermediate step between aftermarket air coolers and water cooling. I would like to have seen how this compared against a high-end air cooler (Turniq Tower, Scythe Ninja and the like). Even a high-end air cooler would cost less than 1/2 of these peltier coolers and it would make no sense to use one of these if an air cooler would provide similar results.

Yeah, that would've been nice. Testing time was limited on this one and I didn't have any good air coolers lying around for reference, but if I could have I would have.

3) Noise. The only reason, other than overclocking, to use a system like this is because of noise. You should have included some graphs with noise measurements of all of the coolers at idle and load.

I tried to describe noise difference in the article. I've used decibel meters in the past but they don't seem to accurately reflect the experience all that well, I prefer to keep this one subjective. Maybe a scone meter would be better, so I'm looking into getting one of those.

Later gents!

 

[qbnsuperman]

i dont understand what is different, I have a freezone on a core 2 quad q6600. I idle at 4C (freezone on high) and under load for about an hour it hits ~14C. I have a lian-li pcv case.. i guess that could be why.

 

[cleeve]

Could be the load you're using.

I used orthos for 10 minutes to get a load temp, Orthos is pretty hard on a CPU.

 

[mr_fnord]

Have you calibrated your temp sensors? 4C sounds a bit too cold for a processor with anything short of phase change cooling. That's 39F on the silicon, and the coolant would have to be at or below freezing in order for the heat to move through the chip packaging and heat exchanger and keep the die that cold. If you are actually that cold you probably have some severe condensation problems.

A lot of motherboards, including the one I'm using right now, have a 20C offset, so your idle temp would be 24C and load would be 34C, still decent, but no risk of condensation. Currently my sensors read 12C on air cooling in a 24C room...

 

[qbnsuperman]

well lol, i'm obviously not as hardcore as you guys; my load was done by playing BF2 and my temps are the mobo temps (as read through pcwizard). I did shoot them w/ an infrared probe and i got 5-6C. Oh, and i do get some condensation on the water blocks that cool the peltiers (blue parts), but only if i turned the a/c off. (we have an A/C contracting company so its always icey cool in here and that = low ambient humidity)

 

[cleeve]

I assume only one is rated for 175 watts. Which one is it and what is the other one rated for? If I'm wrong, what makes one more powerful than the other one?

You're right, the Freezone should be 175 and the Eliminator should be 125.

 

[tleavit]

But how quite is this thing. My problem right now is on my HTPC, the CPU fan under load and in the warmer summer days is louder then a friggen vacuum cleaner and it sits on my theater rack. Man, I jsut want a PC that you don’t hear. In this continued world of increasing PC power, the sound factor keeps getting worse and worse.

 

[gm0n3y]

Glad to hear there is another article coming up where you really test these babies out. Hopefully in that one your can throw in a high-end air cooler for some comparison.

 

[custompcz]

I concur with both posts that this review is lame if your only gettin' it to 2.25Ghz on the OC. At that level of OC, stick with the stock cooler and save your $$!!! If that doesn't work for you, then step up and get a decent HSF for well under a C note.

Pathetic as far as I'm concerned considering my modded Zalman in my sig with "only" ambient temps at best. I have been considering adding a TEC to it and see how that goes.

Trying out the PDC E2160 next to try and emulate X-bit labs success and perhaps exceed the 3.4Ghz limit they hit on air. Has anyone done better than 3.4Ghz on air with the E2160?

 

[custompcz]

Mod yourself a Zalman reserator, you'll be happy with the end result.

 

[gm0n3y]

You could try underclocking your CPU, assuming that it would still be adequate for the job (which, if your considering a $200 TEC, it had better be a higher end CPU).

 

[nukemaster]

It's all good.

I have wondered about how those things where.

Why the hell do people whine about the reviews. If they don't like it, write there own! 😛

 

[pmr]

With its Freezone and the Eliminator coolers, CoolIT brings hybrid Peltier/liquid cooling to the masses. How do the two coolers stand up against cooling systems using other methods?

http://www.guru3d.com/article/processor/438/

 

[yadge]

Edit

Just realized it was a stupid question.

 

[harmattan]

I've had mediocre results with my Freezone. Prior to the Freezone, I was running a Scythe INF air, 50c idle/57c load. I'm now at 48c idle/56c load with the Freezone, so a net improvement of 3c.

The article touches upon the main problem with the Freezone and Eliminator which is, once the cpu has run load for a while (30 minutes or more in my case), the amount of water in the resevoir and length of the loop cannot cool water fast enough.

Also, this thing is L-O-U-D. Installation was also a hassle: the 120mm fan adapter doesn't match holes on the TT Armor case and some drilling is necessary.

 

[dragonsprayer]

THG has done once again what a horrible review!

I just happened to rip out one of these piece of junk freezone - what junk!

noise!

70-80c vs 45-50c for koolance 1000 system

total junk!

yes it was 3.6ghz quad core but who would buy a $300 cooler for $100 e4300 cpu?

what a junk article!









this cooler does not even come close the 175w cooling claim! how can something that is so inefficient only draw 50w and cool 175w?

it can not - its obvious since i tore the junker out!

 

[dragonsprayer]

Could be the load you're using.

I used orthos for 10 minutes to get a load temp, Orthos is pretty hard on a CPU.

koolance 1000 runs orthos all day long at 50-55c

this freezone piece of junk! craps out after 2min crashing the system at 75c or higher.

sometimes it even goes nuke - it must be due to air lock? and the system hit 80c on idle what junk!


any one want to buy one cheap email me!

 

[Yoosty]

Good article. It's interesting to see that plain old water cooling is actually better under load than Peltier-chilled water setups. Of course, part of this could be the prepackaged Peltier cooling unit vs the large, customizable water setup. Are there single-unit water setups like the Freezone and Eliminator, only without the Peltiers?

In other words, does the worse cooling performance of the Freezone come from the very limited size of the unit, or are the Peltiers just inefficient/not as effective at transferring heat as a plain old radiator? Fighting the heat inside of the case doesn't help I'm sure, but it would be interesting to see where the bottlenecks are, so to speak.

Yes there is. Cooler Master Aquagate Mini R80 80mm Water Cooling Kit is a self contain system and a POS.

 

[Farrwalker]

I have been useing a Freezone in one of my computers for over a year now. It is cooling an overclocked Opteron 165 (1.8GHz std) clocked at 2792MHz. It is extremely quiet. The only noise is the fan and that is very quiet. Under load of Prime 95 running on both cores it goes up to 58C, if I remember correctly (can't find my yellow pad that I wrote my notes on from a year ago).

If I overclocked any higher the temperature just would creep higher and higher; the Freezone is unable to handle any more heat generation. I even tested useing a much higher capacity fan and it only helped a little; but the noise of a 120CFM fan was intolerable for a couple of degree improvement.

Motherboard: DFI nF4 SLI-DR Expert
CPU: Opteron 165 CCB1E 0609FPMW @2792
Memory: 2 x 1GB OCZ DDR 500
Graphics card: XFX 7900GTX 512MB
PSU: Sparkle FSP650
CPU Cooling: Cool It - FreezeZone (thermo electric and liquid cooling)