Koolance's Response to Hardware Labs "The Copper Radiator Myth"

[migueld]

ThermoChill's Official Response...

We have no problems with Koolance's data - it is accurate. We do however question it's usefulness to the market at which our products (and theirs) are aimed and obviously question the suitability of the parameters & equipment used for their testing. We would like to formally suggest that Koolance provide a sample of their radiator featured in that test to be submitted to an independant tester so that it can be subjected to tests under parameters that are applicable to our scene, using only fans used within this market, and flowrates and pressures reproduceable by the pumps within this market. This testing should be done by an established independant neutral source using established testing methodology.

We would also like to thank Koolance for demonstrating that even under such ridiculous testing conditions, our PA Series radiator outperforms the HWLabs GT Stealth 240 radiator in both Hydraulic Resistance and Thermal Capacity.

Wisest post so far.

[NaeKuh]

Oh god. Here we go again. Top Nurse won't give up on [H] and now she's at it here with more BS. Give the agenda a rest. Koolance is a joke along with their use of aluminum and bogus testing techniques.

Goes to show how honorable and ethical a company is to provide and promote misinformation. It's amusing to see which Koolance lackeys are enamored with Koolance's response wishing it were true.

ranker... lets be nice for once. Top nurse hasnt said anything offending.

And please read cathars post. They need to use fans of real numbers and coolant on REAL numbers.

There is no way in hell my cpu runs at 85C. If my coolant ran that hot, id be back on air.

And everyone knows radiators are built with specific CFM and Heat loads. A good example is a PA120.3 with low fans pwning all the other radiators out there.
And a BIGTX with high velocity fans. I wonder how a HA would of done on those tests.

[Cathar]

That's not true. Koolance explicitly mentions all three components of a brass radiator: brass tube, bond and copper fins, AND mentions copper's properties: "Since copper has a thermal conductivity 1.8 times greater than aluminum..."

...go on, why didn't you quote the full sentence from Koolance?

"Copper Core" radiators are touted as offering the highest performance for liquid cooling systems. Since copper has a thermal conductivity 1.8 times greater than aluminum, a copper radiator should outperform aluminum, right? Actually, this is false.

They specifically omitted discussional analysis of the long & short heat paths. The deliberately played up the aspect of the short brass path, and chose not to make any further mention of the long copper path. THAT is omission.

I could care less what HWLabs wrote. One does not gain credibility by sinking to the same depths as someone else's murky marketing.

[DepTi]

All very interesting so far. Great posts Cathar and Marci.

[ColonelCain]

Yes. Great posts Cathar and Marci.

Whenever a bullsh** company such as koolance releases such ridiculous claims such as this, the end result is their "results" is pulled and ripped apart to expose it's weakness's by people such as ourselves.

[hwlabs]

Ironically, HWlabs is the one conveniently omitting information in their analysis, like I mentioned some time ago: they don't specify the brass heat transfer properties nor they specify the solder's properties. That is a big omission.

Koolance mentions that design plays an equally important role in radiator performance. I'm in complete agreement. I think the PA was successful because of the design rather than the elements. So, far from being an alu vs brass argument, I consider this being an informative debate. Sure there is hype and marketing, Koolance is guilty of that, and HWLabs is guilty as well. But both have worthwhile data, and I think Koolance has partially redeemed itself with this analysis. I just hope they don’t turn into arrogant pigs

I beg to disagree, miguelD.

There is no need to contend with an accepted scientific fact nor omit it.

We've made it rather clear that even if brass does have a lower conductivity, the thickness of the tube walls and even the solder material is still far thinner than that of the total tube-fin material employed in an Aluminum radiator.

Koolance was the one who brought up the idea of oxidation, which is naturally occurring, and mentioned it was a factor for retarding performance. We simply told them that if there is any such impact, it would be negligible.

What Koolance simply did was showed a scenario wherein its heat exchangers performed well enough to prove its self-serving assertions while conveniently omitting the fact that such operational conditions, i.e. air velocities of 5m/sec are inapplicable to PC watercooling.

We did mention in our open letter to them that:

The "aluminization" of the Automotive industry was brought about by cost cutting regimes implemented by various automotive manufacturers. Aluminum is a very good alternative for this solution provided it is given sufficient frontal surface area and high airflow. Both parameters which are unfortunately luxuries that PC cooling do not have.

Even after reading this, they completely ignored the critical operating parameter and went on with their tirade anyway.

The noise generated by such a fan would require you to keep your PC in an acoustically insulated cabinet or its own room.

Just the same, what Koolance did successfully highlight was the fact that the GTStealth core design will perform to the levels of a two-row radiator. But this is besides the point, as radiators like the TC PA's were designed primarily for a specific application.

Notice the fact that their PR material is beginning to subtly drift away from PC radiators and dipping itself to purely HEAT EXCHANGERS which would clearly suggest the fact that selective PR spin is in effect.

Design does play an important role in radiators. You design according to what you want your radiator to perform under certain operating conditions. Our radiators are used in a host of applications related to PC watercooling and beyond which would require more capacity and compression. Thus they were designed as such.

Supposing you do give us 5m/sec of air velocity to work with, Marci/Cathar, we do have a platform that would easily outperform the PA's at half the size.

Now if Koolance wants us to design a heat exchanger for use under automotive applications, then all they're doing is preempting one of our development programs. But I do recall their market to being one focussed only on PC water cooling.

Now I do wonder why they've chosen not to include the GTX240 in their tests. Our industrial radiator designs have shown our two row radiators to outperform 5-6 row units but that is of course not as relevant to the PC market given the specific operating parameters.

However, don't ever think we're sitting on our laurels. There are distinct advantages to not relying on a sub-contractor to do your engineering and manufacturing work for you. It costs a substantial sum for a set of tools to generate each new fin geometry. Speaking of which, they will be arriving soon.

[NaeKuh]

I beg to disagree, miguelD.

There is no need to contend with an accepted scientific fact nor omit it.

We've made it rather clear that even if brass does have a lower conductivity, the thickness of the tube walls and even the solder material is still far thinner than that of the total tube-fin material employed in an Aluminum radiator.

Koolance was the one who brought up the idea of oxidation, which is naturally occurring, and mentioned it was a factor for retarding performance. We simply told them that if there is any such impact, it would be negligible.

What Koolance simply did was showed a scenario wherein its heat exchangers performed well enough to prove its self-serving assertions while conveniently omitting the fact that such operational conditions, i.e. air velocities of 5m/sec are inapplicable to PC watercooling.

We did mention in our open letter to them that:


Even after reading this, they completely ignored the critical operating parameter and went on with their tirade anyway.

The noise generated by such a fan would require you to keep your PC in an acoustically insulated cabinet or its own room.

Just the same, what Koolance did successfully highlight was the fact that the GTStealth core design will perform to the levels of a two-row radiator. But this is besides the point, as radiators like the TC PA's were designed primarily for a specific application.

Notice the fact that their PR material is beginning to subtly drift away from PC radiators and dipping itself to purely HEAT EXCHANGERS which would clearly suggest the fact that selective PR spin is in effect.

Design does play an important role in radiators. You design according to what you want your radiator to perform under certain operating conditions. Our radiators are used in a host of applications related to PC watercooling and beyond which would require more capacity and compression. Thus they were designed as such.

Supposing you do give us 5m/sec of airflow to work with, Marci/Cathar, we do have a platform that would easily outperform the PA's at half the size.

Now if Koolance wants us to design a heat exchanger for use under automotive applications, then all they're doing is preempting one of our development programs. But I do recall their market to being one focussed on PC cooling.

Now I do wonder why they've chosen not to include the GTX240 in their tests. Our industrial radiator designs have shown our two row radiators to outperform 5-6 row units but that is of course not as relevant to the PC market given the specific operating parameters.

However, don't ever think we're sitting on our laurels. There are distinct advantages to not relying on a sub-contractor to do your engineering and manufacturing work for you.

This is an easy answer. Because the GTX which is more optimized for higher velocity fans would of kicked the living crap out of that test. As well as the Thermochill HA series radiator.

But to the forum HWlabs! Its always a win on our side when a vendor decides to pop an account and join us!


At anyrate, WHO realisitically runs there coolant at 85C??


Offtopic: can you give us any earily info on your RXN series? Im really interested in your evap cooling scheme. If i recall it was given up on the avation industry because it was fragile in wartime. ??

[Hard H20]

But to the forum HWlabs! Its always a win on our side when a vendor decides to pop an account and join us!

I agree, welcome Hwlabs, it's good to see a response to this bull that Kool-aid is spoon feeding its fanboy base.

At anyrate, WHO realisitically runs there coolant at 85C??

Isn't it obvious? Koolance does to manipulate testing results to favor thier POS product.

[ColonelCain]

Sweet, now we have the 2 vendors that are competing against koolance on our side.
Welcome HWlabs!

[Top Nurse]

I've just bought some new fans.

Picture of me with new fans.

Can anyone recommend a radiator that's appropriate?

I would like to see koolance conform to the standard of sizes set by other rad companys. 120.X radiators.

I suspect that won't happen anytime soon as their rads are sized to fit their own hardware.

Say a PA120.3 with same blocks and pump running 100% distilled water, and a equal size Koolance radiator with the same blocks and pump, but running a 80/20 mixture? (I doubt the normal 95/5 ratio would be enough to stop AL and CU from eating at each other).

The Koolance supplied coolant is 70/30. Which is why I ditched it in favor of my own concoction of 90/10 HyperLube. Now I use a 98/2 mixture of AC Fluid.

[hwlabs]

The Koolance heat exchangers appear to be standard automotive condensers.

We would also like to thank Koolance for demonstrating that even under such ridiculous testing conditions, our PA Series radiator outperforms the HWLabs GT Stealth 240 radiator in both Hydraulic Resistance and Thermal Capacity.

Marci, you wouldn't' be doing cartwheels if you read this: http://www.koolance.com/technical/my...ab_results.pdf

How the PA radiators "outperform" the GT Stealth radiators aren't clear in their tests.

At 5 m/sec air velocity and 5 l/min liquid mass flow:
The Thermochill 120.2 dissipates 6.25 kW
The Black Ice GTS240 dissipates 6.13 kW
The Koolance Radiator dissipates 9.62 kW

But at 10 m/sec air velocity and 5 l/min liquid mass flow:
The Thermochill 120.2 dissipates 7.11 kW
The Black Ice GTS240 dissipates 7.14 kW
The Koolance Radiator dissipates 10.41 kW

And at 3 m/sec air velocity and 5 l/min liquid mass flow:
The Thermochill 120.2 dissipates 5.64 kW
The Black Ice GTS240 dissipates 5.67 kW
The Koolance Radiator dissipates 8.86 kW

The only thing it clearly shows is how much bigger the PA's are how inefficient it is for its size in relation for this wind tunnel application. It also indicates several working points that show very minimal differences between the GTS240 and the PA 120.2 switching favors among the two.

There are some things that remain unclear with how these tests were conducted. The air pressure drop figures in relation to the W dissipated are very incongruent to real world data. How can a radiator of practically exact same dimensions, having a significantly higher pressure drop characteristic, yield 50% more performance? It doesn't make sense.

Somehow I wonder why those test results aren't allowed to be published for legal or advertising or lawsuit, deed etc..?

[DarthBeavis]

You all are just lucky they did not include a Thermaltake radiator in that test . . .












LOL. Maybe pair it up with a Packard Bell uber rig

[Spart]

I'd sign up to test this rad and compare it to my lovely PA120.2............

But only if I was stupid enough to run their POS with my DD Copper Blocks.

[NaeKuh]

You all are just lucky they did not include a Thermaltake radiator in that test . . .












LOL. Maybe pair it up with a Packard Bell uber rig

You must be old to know that name. LoL.

They changed names to HP if i remember correctly. But packard Bell machines were awesome back in the days. Much better then sony vaios, and compaq's that were running around.

Only the IBM machines would be the ones to go over.


This was back in the days when circuit city was fighting good guys for electron store superiority. And when AOL chatting was hella addictive.


But onto a different note:
I want to hear more about the RXN!! To me, i have no loyalties to any company, except vendors. Like petra's and jab-tech. If swiftech has the best block on the market, i buy it because its the best, not because gabe is a great guy. Which he is.

Same goes with thermochill and marci. I buy there products because there crowned the best. If the RXN can prove to be a better radiator, me and about 1/3rd of the forum would be all over it.

So what exactly is the RXN? can you tell us something about it, without breaking the NDA you have? Oh and a lot of people are wondering if the girl is either a real model or computer generated.

[hwlabs]

OT: We still can't reveal the details on the RXN just yet. The consortium involved in developing this technology will need to be in consensus to release more technical information on the RXN. Sorry naekuh .

But suffice to say, the model was real, she's a nice lovely lass named Kathy Moya. The photo shoot was fun albeit tiring but the concept was all that we deemed it to be. The updates will only get hotter from there.

[p8ntslinger676]

I would love to see a real "no nonsense" third party test, meaning the copanies put up their best 120 x 3 rad and see how well they really do with a standard heat load, (say 500watts) and a assortment of different fans starting with low flow yates and making your way all the way up to delta's now that kind of test would really set the rads apart and i also have a feeling that koolance is trying to pull some as they normally do. Im pretty sure that if you did that you would come to pretty conclusive results that overall the hwlabs and thermochill rads greatly outperform the koolance rads

[IanY]

You must be old to know that name. LoL.

They changed names to HP if i remember correctly. But packard Bell machines were awesome back in the days.

NaeKuh,

Huh? You're obviously not old enough

Packard Bell was started in the 1920s and was bought out in a LBO by private investors in 1968 from Teledyne. Packard Bell has *nothing* to do with HP or Bell Labs (now Lucent/Alcatel). In the mid-1990s, Packard Bell acquired Zenith Data Systems and that sparked the interest of NEC. Packard Bell was then majority owned by NEC (and minority owned by private investors) and kept as a NEC brand name. HP got into a tiff with NEC over the continued use of the Packard name, but they gave up after a while. NEC got smart and eventually got out of the PC business in the US, with a flood of crap computers like eMachines. The Japanese, after a couple of painful years, eventually got smart and, in 2006, thankfully sold Packard Bell to the former owner of eMachines, a Chinese guy who sold the crap eMachines brand to Gateway sometime in 2004.

How I know this? I'm in the LBO business. Besides, this crap should be on Google.

So, Packard Bell has nothing to do with Hewlett Packard.. thankfully.

[Sparky]

I've just bought some new fans.

Picture of me with new fans.

Can anyone recommend a radiator that's appropriate?

rofl

[nikhsub1]

But onto a different note:
I want to hear more about the RXN!!

I think the RXN is a self contained Thermo-siphon type of deal - think asetek vapor chamber: http://www2.asetek.com/main/page.asp?sideid=623
This is my guess anyway.

[SDatl404]

Go Koolance! All the people here talk smack and pull out all there calculators trying to find reasons to shoot holes in the tests.

Enjoy the Koolance ownage!

Until someone proves koolance inferior with tests, Koolance is now the "king of cool"!

[Cathar]

On the subject of touting efficiency for size for PC watercooling radiators

The only thing it clearly shows is how much bigger the PA's are how inefficient it is for its size in relation for this wind tunnel application. It also indicates several working points that show very minimal differences between the GTS240 and the PA 120.2 switching favors among the two.

In all published documentation on the PA radiators, it has always been disclosed that the PA radiators were not designed to operate well at high air velocity speeds, and certainly not in high-powered wind-tunnel applications. This level of disclosure went so far as to actually recommend the use of different radiators if fan speeds were above a certain point. In fact, I'm somewhat surprised that the PA performed as well as it did, operating so far outside of its targetted operational parameters.

The size of the PA's was dictated by the maximum performance achievable given typical PC fan cooling powers. Performance for size was NOT a primary design concern, and it clearly shows. The only thing that was important was performance. Following that path brought the design to the point that it is now at.

I tend to view companies that go out of their way to stress "performance for size" as missing the point, but dredging this point up in some vain way to claim any measure of superiority for purposes of the intended target application. If the design goal was "peak performance within a certain core thickness", you can bet your bottom dollar that the radiators would look different.

Oddly enough, this differentiation, core thickness, is of minimal concern for enthusiasts who want the best cooling power on the market for low noise fan use. An extra 2-3cm of radiator depth is of minimal concern when the PC water-cooler only wants the lowest temperatures.

It's not like the car auto industry where performance for size is a major concern and design consideration. I believe that touting this point belies a misunderstanding of the target market by any company that continues to apply auto-industry design criteria to the PC water-cooling market, and to attempt to use that to claim as any measure of technical superiority is merely clutching at straws. In a car, performance-size(weight) is important. In a PC, far less so.

[nikhsub1]

Yup, koolance ownz, we are all SUCKERS! ROFLMFAO.

[SDatl404]

Yup, koolance ownz, we are all SUCKERS! ROFLMFAO.

They proved it... What did your company do?

[hwlabs]

That is why we are in agreement that the Koolance tests and its operating parameters are inapplicable, Cathar. It was Marci who was thanking them for Thermochill rads showing superior(?) results over the Black Ice GTStealth when said results would show otherwise.

That is also why as a policy, with the exception of Koolance, we don't play this sniping game. Its simply unethical.

[nikhsub1]

They proved it... What did your company do?

Since you obviously can't read - this post will be of little help to you but I'll do it anyway. In koolance's particular test, they ran the coolant at 85c and airflow at ~320CFM LOL. Now, if they performed this test in an environment that we actually see in a computer (not in a car - that testing was done automotive style) which would be something like 30c coolant temp 23C ambient temp and ~80 - 150 total CFM through the rad. I assure you results would be far different.