MMM
Page 1 of 11 1234 ... LastLast
Results 1 to 25 of 259

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

  1. #1
    Xtreme Enthusiast
    Join Date
    Jan 2007
    Location
    Coeur d' Alene, Idaho, USA
    Posts
    585

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

    Me thinks HWlabs bit off a little more than they could chew!

    HWlabs Open Letter to Koolance


    Koolance's Response to Hardware Labs (June 15, 2007)

    The Copper Radiator Myth

    "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.

    By itself, copper does indeed transfer heat more quickly than aluminum. This is why most liquid cooling companies, including Koolance, base water blocks in pure copper. But one physical property of a material does not dictate performance under every application. This is particularly evident when dealing with extremely thin materials, such as radiator walls.

    Thin strips of copper can not be welded or brazed without melting. That means a louver fin radiator that claims to be "all copper" or "copper core" actually consist of multiple materials You would expect a "copper core" radiator to directly pass heat from liquid into copper. However, this radiators does not. To get a better idea of what's going on, you need to split the radiator open. (Don't worry, we'll do this for you)
    Here we find this radiator to be predominantly brass. Liquid never comes into contact with pure copper, which is only used in the air fins. Therefore, this technique requires a third material to join the two metals, which means the heat transfer efficiency is significantly reduced.

    Furthermore, typical brass has a thermal conductivity of about 159W/mK (37/15 Cu/Zn). Koolance uses nearly pure aluminum in its radiator fins and tubes, which has a thermal conductivity of about 237W/mK (33% greater).

    Based on these points, it should not be difficult to realize that an all aluminum heat exchanger will actually transfer heat more quickly than a mixture of different materials




    In extensive testing, Koolance has found our all-aluminum brazed radiators transfer heat 39-49% more efficiently than a typical claimed "copper core" radiator (per cubic centimeter of radiator).

    So why would any company utilize this brass/copper radiator design? The simple reason is cost; this manufacturing technique is cheaper. Brazing a full aluminum heat exchanger requires a relatively large amount of initial expense. (See notes on corrosion.)

    Similar to abandoning cheaper tube and fin copper radiators, the automotive industry long ago transitioned to all-aluminum brazed radiators for their higher level of performance.
    Koolance's Response to Hardware Labs





    Testing Console at KATECH


    Koolance created a stir last month when we posted that HWLabs Black Ice® radiators utilized a lower-performing design to our own. It was not our original intention to point out specific radiator manufacturers. In fact, the Copper Radiator Myths page did not list any company brands or names until later, when customers requested this information. To preface this page, Koolance has found nothing wrong with the quality of HWLabs radiators.

    However, after submitting samples to an independent laboratory (KATECH), we are maintaining our original argument. We also included a ThermoChill radiator during this lab analysis. The Koolance brazed full-aluminum radiator outperformed both tested brass/copper radiators relative to size. Full test results have been posted here.

    Copper conducts heat more quickly than aluminum, and Koolance has never asserted otherwise. Some individuals have made the argument that this alone means a copper radiator will outperform aluminum, but that reasoning disregards the fabrication of those materials. A hypothetical heat exchanger incorporating carbon nanotubes would not necessary perform better than aluminum or copper simply because of its composition. The design matters equally!

    Since HWLab's reponse, some have accused Koolance of spreading disinformation, or wrongfully questioning the "industry standard". Notwithstanding the fact we feel we're correct, where would technology be if no individuals or companies were willing to question standards? Koolance has posted this information because we felt it beneficial to discuss given the opposing viewpoint (which we feel is a general misconception).

    To address HWLab's "Erroneous" assertion points:

    1. "Glue is used to bind the fins to the tubes..." Koolance originally used the term "glue" improperly due to a language mistranslation. This was intended to indicate the effect of the bonding material, and not the actual material itself since various substances and techniques can be used by different companies. Our original point that a third material is used to bond copper fins to brass tubes still stands-- whether it be solder, metal powder in an adhesive, or another technique.

    2. "Brass tubes in copper radiators make for poor performance..." Both radiator designs are valid, we simply stated that brazed aluminum performs better relative to size.

    HWLabs here failed to comment on the thermal performance of brass, which is the only radiator material in contact with water. Typical brass has a thermal conductivity 33% lower than aluminum. So taking only relative thermal conductivities into consideration, copper is about 70% faster than aluminum, and aluminum is about 50% faster than brass. HWLabs also mentions materials thickness, and herein emphasizes our very point: these metals are less critical when such thin pieces are used.

    Considering these aspects, Koolance's argument for higher performance with a 100% aluminum radiator over a brass/copper amalgamation should appear feasible.

    3. "Copper is prone to oxidation and re tards performance..." Copper and aluminum both oxidize (as Cu2O and Al2O3, respectively), and each can help protect from corrosion. How significantly this affects performance is subject to debate. However, HWLab's assertion that oxidized copper is "not as dangerous" as a pinhole cavity in aluminum appears incongruous. [In an unrelated note, aluminum better protects some metals like brass from corrosion when alloyed in small amounts.]

    The dissected Black Ice® heat exchanger (WT-Xtreme-BK) was not the identical one tested (WT-GTS240). It was actually purchased new from a U.S. reseller, and water was never run through it. Therefore, contrary to HWLab's implication, Koolance did not "accelerate" oxidation, and we would have to assume someone who dissected that particular model of radiator would encounter similar results. It is certainly possible the unit was used before we purchased it, or else attributable to a manufacturing anomaly.

    4. "Aluminum is always the best..." What Koolance maintains, is that an all-aluminum brazed heat exchanger of identical size as a brass/copper heat exchanger will have a higher heat dissipation potential. The lab's results demonstrate this.

    The required initial investment in manufacturing equipment is normally greater for aluminum heat exchangers than brass/copper. Naturally, the per unit cost for either radiator design can be reduced with sufficient production.

    HWLabs indicated that [copper] round tube and fins have not been used in automobiles for over 50 years. Koolance did not describe the tube and fin heat radiator in detail on the myths page. We linked to a separate page describing the disadvantages of this design because some PC cooling companies still utilize this technology. It is not strictly relevant to Koolance's base argument for aluminum versus brass/copper louver fin style heat exchangers.

    Koolance is not bound by any particular heat exchanger technique. If we decided that aluminum's advantages no longer outweighed an alternative, we would change for the benefit of our customers! We are already in the process of increasing the thickness and flow rate of our heat exchangers to further their performance over brass/copper.

    Heat Exchanger Lab Results

    Radiator Design

    Heat Exchanger Lab Results (Continued)

    The Copper Radiator Myth

    KATECH

    Moving Heat
    "If It Doesn't Overclock Its Broken"
    "NexGen Overclocking"
    "Overclocking, its a life style"

  2. #2
    Banned
    Join Date
    Jun 2005
    Posts
    197
    Interesting reply and backed up with outside testing to boot.

    I shall have to get one of these radiators and compare it to the results I'm getting now with a HWL stealth now. Just got to figure out how to mount it to my AC grill.

  3. #3
    Xtreme Addict
    Join Date
    Feb 2007
    Location
    Arizona, USA
    Posts
    1,700
    Well, if you take this into consideration, then have them prove why all watercoolers aren't using koolance rads?


    Core i7 920 D0 B-batch (4.1) (Kinda Stable?) | DFI X58 T3eH8 (Fed up with its' issues, may get a new board soon) | Patriot 1600 (9-9-9-24) (for now) | XFX HD 4890 (971/1065) (for now) |
    80GB X25-m G2 | WD 640GB | PCP&C 750 | Dell 2408 LCD | NEC 1970GX LCD | Win7 Pro | CoolerMaster ATCS 840 {Modded to reverse-ATX, WC'ing internal}

    CPU Loop: MCP655 > HK 3.0 LT > ST 320 (3x Scythe G's) > ST Res >Pump
    GPU Loop: MCP655 > MCW-60 > PA160 (1x YL D12SH) > ST Res > BIP 220 (2x YL D12SH) >Pump

  4. #4
    Xtreme Enthusiast
    Join Date
    Jul 2006
    Posts
    620
    hope the mods keep this thread on a short leash
    “You Americans are so gullible. No, you won’t accept communism outright, but we’ll keep feeding you small doses of socialism until you’ll finally wake up and find you already have communism. We won’t have to fight you. We’ll so weaken your economy until you’ll fall like overripe fruit into our hands." Nikita Khrushchev

  5. #5
    Xtreme Member
    Join Date
    Dec 2006
    Posts
    393
    Whoa, what a H-Bomb. That was unexpected. It's nice to see Koolance doing their research though. I had never seen any data comparing a Koolance radiator vs brass-based.

    I don't know if their research is valid, but if anyone wants to bash Koolance I'd suggest doing the same kind of testing. Buy their thing, do unbiased testing and talk. More baseless bashing won't get us anywhere.

  6. #6
    Xtreme News
    Join Date
    Dec 2005
    Location
    California
    Posts
    1,607
    Quote Originally Posted by migueld View Post
    Whoa, what a H-Bomb. That was unexpected. It's nice to see Koolance doing their research though. I had never seen any data comparing a Koolance radiator vs brass-based.

    I don't know if their research is valid, but if anyone wants to bash Koolance I'd suggest doing the same kind of testing. Buy their thing, do unbiased testing and talk. More baseless bashing won't get us anywhere.
    Agreed. If your going to bash, we will need some proof

  7. #7
    Xtreme Enthusiast
    Join Date
    Sep 2006
    Location
    brazil
    Posts
    829
    Water Temperature: 84(± 1)°C

    I know they went with "extreme" temps for testing purposes, but 84c? even tygon is rated at +- 70c... would this alone favor alumminum rads?

    E8600 - Maximus Formula II - HIS 4870x2 - Mushkin DDR2 1066 2x2GB - Silverstone DA 850W - Silverstone TJ09 - TRUE Black

  8. #8
    I am Xtreme
    Join Date
    Aug 2003
    Location
    Cherry Hill, NJ
    Posts
    8,291


    Can someone explain that?
    Phenom II 940 BE / ASUS M4A79 / HD5770 Crossfire
    3770mhz CPU 2600mhz NB | DDR1040 5-5-5-15 | 900/1250

  9. #9
    Banned
    Join Date
    Jun 2005
    Posts
    197
    Quote Originally Posted by ColonelCain View Post
    Well, if you take this into consideration, then have them prove why all watercoolers aren't using koolance rads?
    Perhaps it's just a simple misunderstanding of information. Like the bombshell Cathar dropped recently on big ass tubing.

    However, I should point out that I ran a Koolance Exos 1 for over four (4) years without a hitch and it has an aluminum radiator in it. I thought it worked great.

    However, to do significant testing in my loop I need to make sure that we aren't comparing apples and oranges. IOW, I need to make sure we are comparing similar radiator sizes and IIRC the Koolance radiators are different in physical size.

  10. #10
    Xtreme Member
    Join Date
    Jun 2005
    Location
    United Kingdom of Great Britain and Northern Ireland
    Posts
    464
    Something I couldn't make hear nor tail of is the air velocity. I have no idea if that's a draft, or if it's a hurricane. And yeah, the 81 degrees is a bit odd to use.

  11. #11
    Xtreme Enthusiast
    Join Date
    Jul 2005
    Posts
    593
    Quote Originally Posted by afireinside View Post


    Can someone explain that?
    I did not fully read the article, but it could be the total amount dissipated in given time length.
    My heatware: http://heatware.com/eval.php?id=41179

    cpu: e6600
    mb: evga 680i
    ram: Ballistix 8000
    video: XFX 8800 GTS
    pw: 750 Quad

  12. #12
    Xtreme Legend
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    17,405
    well i think we all learned something new today

    looks around for an alu rad to test ROFL


    now that they've done this why not supply their rad for testing to billa and others to see numbers we can easily relate to
    Team.AU
    Got tube?
    GIGABYTE Australia
    Need a GIGABYTE bios or support?



  13. #13
    Banned
    Join Date
    Jun 2005
    Posts
    197
    Quote Originally Posted by Halk View Post
    Something I couldn't make hear nor tail of is the air velocity. I have no idea if that's a draft, or if it's a hurricane. And yeah, the 81 degrees is a bit odd to use.
    I saw that big squirrel cage blower they were using for air flow and wondered just how much air they blew through that radiator. They should have used the same size fans as would have been used in their products.

  14. #14
    Xtreme Enthusiast
    Join Date
    Jul 2006
    Posts
    620
    I saw that big squirrel cage blower they were using for air flow and wondered just how much air they blew through that radiator. They should have used the same size fans as would have been used in their products.




    Instead of just walking away from an embarrassing pr stunt they chose to dig deeper
    Last edited by _G_; 06-16-2007 at 12:34 AM.
    “You Americans are so gullible. No, you won’t accept communism outright, but we’ll keep feeding you small doses of socialism until you’ll finally wake up and find you already have communism. We won’t have to fight you. We’ll so weaken your economy until you’ll fall like overripe fruit into our hands." Nikita Khrushchev

  15. #15
    Banned
    Join Date
    Jun 2005
    Posts
    197
    Quote Originally Posted by _G_ View Post

    Instead of just walking away from an embarrassing pr stunt they chose to dig deeper
    Well the big mistake they made was going to an automobile testing facility. That squirrel fan was probably the smallest fan they had in the place.

  16. #16
    Banned
    Join Date
    Jun 2005
    Posts
    197

    Question

    Quote Originally Posted by hoax View Post
    Water Temperature: 84(± 1)°C

    I know they went with "extreme" temps for testing purposes, but 84c? even tygon is rated at +- 70c... would this alone favor alumminum rads?
    84 C = 183 F What was Koolance thinking about? I wonder if they just manipulated the testing conditions till they got the data they wanted. That temp seems more like what you might see in an automotive cooling circuit...

  17. #17
    Xtreme Enthusiast
    Join Date
    Jul 2006
    Posts
    620
    all they did was "adjust" the test parameters until there product showed good results

    ^^beat me to it lol
    bet it was around 1500 to 2000 cfm too
    “You Americans are so gullible. No, you won’t accept communism outright, but we’ll keep feeding you small doses of socialism until you’ll finally wake up and find you already have communism. We won’t have to fight you. We’ll so weaken your economy until you’ll fall like overripe fruit into our hands." Nikita Khrushchev

  18. #18
    Registered User
    Join Date
    Jun 2007
    Location
    Melbourne, Australia
    Posts
    2
    Quote Originally Posted by _G_ View Post
    all they did was "adjust" the test parameters until there product showed good results


    it shows in some of their cars too.......

  19. #19
    Xtreme Legend
    Join Date
    Nov 2002
    Location
    Melbourne, Australia
    Posts
    1,025
    Completely, utterly, and totally irrelevant.

    Quote Originally Posted by Halk View Post
    Something I couldn't make hear nor tail of is the air velocity. I have no idea if that's a draft, or if it's a hurricane. And yeah, the 81 degrees is a bit odd to use.
    I took the time to calculate the actual air-flows:

    PA120.2 => 324CFM
    HWLabs GT240 => 318CFM
    Koolance => 330CFM

    So, these are all 2x120mm fan radiators. That means to achieve that sort of air-flow, we'd need to be pushing ~160CFM actual, through the radiator, with each fan.

    What sort of fan could do that you ask? 160cfm @ ~110Pa (~0.45in H2O).

    Found one!

    The Delta TFB1212GHE. This lovely little number is a 220CFM fan, running at a delightfully pleasant 65.0dBA. I have one of these fans at home. They are freaking insanely loud. Think of your typical hairdryer, on the highest speed setting, and then turn on 9 more of them, and you're still probably not quite there with understanding just how loud and annoying these fans are.

    Make absolutely no mistake. This is marketing/advertising crap at its total sleaziest worst.

    It is of absolutely zero secret that the design for the PA120 series has its efficiency falling away rapidly with the use of fans of >100cfm rating. Above 130cfm fans the PA120 radiators really don't benefit from any extra airflow at all.

    This is Koolance purposely choosing the "weak-points" in competing designs, and constructing an artificial test that obfuscates the full disclosure of the fan power and noise required to attain the air-flow levels demonstrated in their tests, and then has the audacity to claim technical superiority.

    Once again we see sleazy marketing practices at their worst brought to play against competitors, and targetting the uninformed consumer who will blindly swallow complete tripe all because it "looks official".

    It's nothing more than tarted up pig swill with a pretty ribbon on top.
    Last edited by Cathar; 06-16-2007 at 01:59 AM. Reason: Corrected. Got CFM wrong. Actually undercalculated it originally!!

  20. #20
    Xtreme Legend
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    17,405
    hence my comment to have someone we can trust test it to confirm these numbers.....or in this case deny
    Team.AU
    Got tube?
    GIGABYTE Australia
    Need a GIGABYTE bios or support?



  21. #21
    Xtreme Addict
    Join Date
    Mar 2007
    Location
    Huntington Beach, CA
    Posts
    1,922
    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.
    Last edited by ranker; 06-16-2007 at 01:45 AM.
    MM Extended U2-UFO CYO (Duality front, Standard back, Horizontal Mobo brace) Anodized Black || eVGA X58 || Intel i7 920 || 6 GB Corsair Dominator PC3-12800|| eVGA 295GTX || Asus Xonar Essence STX || VisionTek 650 TV Tuner || 1 300GB WD Velociraptor || 1TB WD's Black Ed. || LG 22X DVD-Writer || Lite-On 20x DVD-Writer || Corsair CMPSU-1000HX PSU

    CPU Loop: DDC-2 w/ XSPC Reservoir Top -> Swiftech GTZ -> Thermochill PA120.3 (Push: 3 Scythe S-Flex G)

    GPU Loop: DDC-2 w/ XSPC Reservoir Top -> (Koolance VID-NX295 FC block) ->Thermochill PA120.3 (Push: 3 Scythe S-Flex G)

    Chipset Loop: DDC-2 w/ XSPC Reservoir Top -> Alphacool Silentstar Dual HD waterblock enclosure -> Bitspower Black Freezer eVGA x58 MOSFET-> Bitspower Black Freezer eVGA x58 NB -> Thermochill PA120.2 (Pull: 2 Scythe S-Flex G)

    2 x Dell 2408FPW LCD || Logitech Z-5500 Digital 5.1 || Logitech G9 Mouse || Logitech G15 LCD Keyboard || Logitech Quickcam Ultravision || Sennheiser Headphones

  22. #22
    Xtreme Addict
    Join Date
    Aug 2006
    Location
    St. Louis
    Posts
    1,055
    Wow, they used the most bizarre set of conditions which are in no way similar to real world usage to get their data, good job. Ok wtf, send computer watercooling parts to an automotive tester?

    Those temps ARE what you would see in an automotive circuit because they sent it to an automotive tester. Why not just do the tests yourself for much less the cost and showing real world results.

    From the PDF:
    Airflow 5 meters per second. WTF? That is 10594 CFM!!!
    Last edited by Chewbenator; 06-16-2007 at 01:35 AM.

    i5 750 @ 3.2 Ghz, stock volts on stock cooling
    4GB G.Skill DDR3-1600 7-7-7-21
    Gigabyte P55-DS2
    EVGA GTX260 192shader

  23. #23
    Xtreme Legend
    Join Date
    Nov 2002
    Location
    Melbourne, Australia
    Posts
    1,025
    Oh, just looking into the hydraulic resistances:

    @5LPM, they're claiming that they measured 9.12kPA @ 5LPM, for the Thermochill PA120.2. 9.12kPA is ~0.93mH2O, or ~1.32PSI.

    Come again? 0.93mH2O @ 5LPM.



    Bill Adams measures just 0.125mH2O @ 5LPM for the PA120.2
    Last edited by Cathar; 06-16-2007 at 01:52 AM.

  24. #24
    Xtreme Legend
    Join Date
    Nov 2002
    Location
    Melbourne, Australia
    Posts
    1,025
    Quote Originally Posted by Chewbenator View Post
    From the PDF:
    Airflow 5 meters per second. WTF? That is 10594 CFM!!!
    Eh? How'd you arrive at that?

    PA120.2 is 0.12m x 0.255m, or 0.0306m²

    At 5m/s, that's 0.0306 x 5 = 0.153m³/sec

    Per minute, that's 9.18m³

    1m³ = 35.3 ft³

    So CFM = 9.18 x 35.3 = 324CFM

    Ack! I got my maths wrong above. The fan power is even more obscenely powerful than I first though.

    [Edit: Edited above post for fan power corrections]
    Last edited by Cathar; 06-16-2007 at 01:52 AM.

  25. #25
    Xtreme Addict
    Join Date
    Mar 2007
    Location
    Huntington Beach, CA
    Posts
    1,922
    Quote Originally Posted by _G_ View Post




    Instead of just walking away from an embarrassing pr stunt they chose to dig deeper
    Seriously. And of course all the grossly misguided Koolance supporters suddenly appear out of the woodwork voicing their support for such a test and commenting on the apparent validity of it.

    So sad...

    *Hands a shovel to the Koolance supporters* Your company's gonna need some help digging with this whopper...
    MM Extended U2-UFO CYO (Duality front, Standard back, Horizontal Mobo brace) Anodized Black || eVGA X58 || Intel i7 920 || 6 GB Corsair Dominator PC3-12800|| eVGA 295GTX || Asus Xonar Essence STX || VisionTek 650 TV Tuner || 1 300GB WD Velociraptor || 1TB WD's Black Ed. || LG 22X DVD-Writer || Lite-On 20x DVD-Writer || Corsair CMPSU-1000HX PSU

    CPU Loop: DDC-2 w/ XSPC Reservoir Top -> Swiftech GTZ -> Thermochill PA120.3 (Push: 3 Scythe S-Flex G)

    GPU Loop: DDC-2 w/ XSPC Reservoir Top -> (Koolance VID-NX295 FC block) ->Thermochill PA120.3 (Push: 3 Scythe S-Flex G)

    Chipset Loop: DDC-2 w/ XSPC Reservoir Top -> Alphacool Silentstar Dual HD waterblock enclosure -> Bitspower Black Freezer eVGA x58 MOSFET-> Bitspower Black Freezer eVGA x58 NB -> Thermochill PA120.2 (Pull: 2 Scythe S-Flex G)

    2 x Dell 2408FPW LCD || Logitech Z-5500 Digital 5.1 || Logitech G9 Mouse || Logitech G15 LCD Keyboard || Logitech Quickcam Ultravision || Sennheiser Headphones

Page 1 of 11 1234 ... LastLast

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •