1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Cooling So what's the current king of air cooling?

Discussion in 'Hardware' started by LennyRhys, 24 Feb 2014.

  1. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    Started out buying everything. Over the years have got to know a few people in the industry who give me something to check out for them from time to time. Still buy most of what I use. I tell it like it is. If there are problems or things that can be improved I say so. Like Thermalright fan clips are the Devil's own until you've used them many times.. and even then far from the easiest to use. But they cool very well and are reasonably priced on the east side of Atlantic. Too bad it's not the same in America. I'm not into it big time but do like knowing what's happening with coolers and fans and which are good all around performers.

    I'm seeing coolers that don't respond to super high airflow as much as they used to. Even with TY-143 fans cooling to airflow starts flattening off around 1800-2000rpm on most.

    I agree the TY-140 is a very good fan and very quiet below 1100rpm.. not very loud even at 1300rpm. TY-143 gets much louder as rpm increase. But it's 130cfm@2500rpm not 74cfm@1300rpm. Using PWM control on all system fans makes a huge difference in cooling performance. I often use both CPU and GPU PWM signals for case fans. This allows for a very quiet system under normal use and still having lots of airflow when needed for extreme loads.

    The size of all larger coolers and pipe spacing means they benefit from 140mm fans so I prefer using 140mm fan.

    Indeed, the 920 is a very good test platform.. Lots of testing having been ran on them makes it easier to compare other testing too. We can compare everything from Ultra-120A from 2007 to newest releases because some of us are still using them.
     
  2. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    Time to get this thread back online as my proposed testing draws nearer...or at least I would like to think so.

    I've been reading all the reviews I can find on the latest coolers and it looks like a really good comparison would be between the Corsair H110 and the Thermalright Silver Arrow IB-E Extreme now that the latter is readily available in the UK (but not cheap at £65).

    I'm surprised to see just how well high-end air performs compared to the latest AIO coolers (at least one review puts the NH-D14 ahead of the H105), and it'll be interesting to see how such capable coolers compare with a very high TDP processor like the i7 920 and a variety of fan types and speeds.

    Incidentally, I found this on ebay today - a 140x140x50mm Delta server fan, a must for any serious enthusiast who uses 140mm cooling haha :D It's rated at 24v but will operate from 8v and, best of all, it's 4-wire PWM. I think I'll bag two for testing purposes.

    [​IMG]
     
  3. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    Nice find!

    Have you noticed the True Spirit 140 Power? It has 6x 8mm heat pipes instead of the usual 6x 6mm heat pipes and is only £35.00. Not many reviews but so far it's looking very good.

    First review is out. 1.3c warmer than NH-D14 2011SE with single fan on 3690X@4.5GHz. Wonder how it will do with 2 fans.. or 2 TY-143 fans?
    http://www.xtremesystems.org/forums/showthread.php?288762-Thermalright-TRUE-Spirit-140-Power

    Another review over on Awardfabrik
    http://awardfabrik.de/guenstiges-trio/10/
     
  4. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    Yeah I saw that today but I didn't realise it has 8mm heatpipes. I was too busy snarling at the inclusion of a 1300rpm fan... why, Thermalright? Maybe they'll release an "extreme" version with one or two TY-143s - now THAT would be killer.

    Speaking of some nice direct comparisons between high-end air, AIO and fully fledged H2O, I found this nugget of solid gold: an apples-to-apples comparison between the H90, H110 and SB-E, also including a 360 EK loop for comparison.

    Amazingly the SB-E and PH-TC14 keep up with the others but the high TDP test separates the men from the boys and the PH-TC14 drops out. From another similar test, the copper TRUE puts up a very good fight and is only a couple of degrees behind the Silver Arrow SB-E.

    Rather than spend £160+ on the IB-E and H110 myself, perhaps it'd be better to wait for poparamiro to add the new Silver Arrow to his catalogue of reviews. I have a strong feeling that the IB-E will beat the SB-E, especially in the high TDP tests.
     
  5. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    Great testing. Thanks!
    The EK loop show just how much difference even a low cost (no such thing) H2O system performs compared to CLC. The overall testing does a good job of showing how well top air perform against CLCs'. I saw a review somewhere with NH-D14 and other top coolers all using high cfm fans. It stays well up in the pack.

    I have Silver Arrow IB-E Extreme now but haven't got it mounted... and when I do I need to OC my 920.. which I have almost no knowledge of how best to do it. My 14PE has TY-143s' on it as do both Silver Arrows. I need to find a couple good 38mm thick fan for my old Cogage Arrow... just so it looks good on the shelf. :D

    Know anybody who wants a Dimastech Easy V3.0? Got one I don't need.
     
    Last edited: 1 Apr 2014
  6. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    The thing with these coolers is that they have incredible heat-dumping potential that is only realised with high CFM fans. In all fairness I think that the real-world difference between the NH-D14, Phanteks PH-TC14 and the Silver Arrow SB-E is negligible, and indeed the difference between these and the best AIO is similarly inconsequential.

    From what I can see, the only real benefit of AIOs is the lack of bulk around the CPU socket meaning that they can be installed in full tower or mATX cases alike.

    Regarding your i7 920, it should be very easy to OC to 4.2 as you have a nice board and 200x21 should be a cinch. One thing's for sure... you don't have to worry about temps, LOL.
     
  7. TaRkA DaHl

    TaRkA DaHl New Member

    Joined:
    15 Mar 2011
    Posts:
    1,635
    Likes Received:
    146
    Lenny... I have a NH-D14 you can borrow to test if you like. Might be a pain in the ass to find any other mounting hardware other than 1155 but I can look if interested. No stock fans either, they're shite. Using AP14s myself.
     
  8. Nexxo

    Nexxo Queue Jumper

    Joined:
    23 Oct 2001
    Posts:
    33,626
    Likes Received:
    1,275
    Spec sheet says it runs from 14V, but it may manage on less. 1.53A, 36W (that's one hungry fan!), 3600 RPM, 273.09 CFM at 61.7 dB. Basically you can build a hovercraft with it, but it will sound like one.
     
  9. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    Very generous offer and yes, that would be awesome - thanks. Incidentally I was looking around yesterday to see if the D14 will mount with the Thermalright 1366 BTK and it looks like it will be fine, so no mounting kit necessary. I'll be getting another TY-143 when I buy the 140 Power so will have plenty 140mm fans.

    @Nexxo, yeah some nice specs there huh :D I'm still toying with the idea of buying a cheap 24v psu because it would be invaluable in situations like this.
     
  10. AlienwareAndy

    AlienwareAndy New Member

    Joined:
    7 Dec 2009
    Posts:
    3,421
    Likes Received:
    70
    Hey I think I have some NHD-14 mounting hardware !

    Deffo got an AMD one if you want that? I keep looking at it and telling myself to throw it away lol
     
  11. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    I have a few coolers too. Let me know if you need one.
     
  12. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    Thanks for the offer doyll - will definitely consider it after the first round of testing.

    @Andy, thanks for the offer - it turns out that TaRkA has the 1366 mount so I now have all that I need to get started.

    The True Spirit 140 Power should arrive tomorrow along with a second TY-143 fan, and if I buy a 140mm Delta today I should have it by the end of the week. Riveting stuff lol. :D
     
  13. AlienwareAndy

    AlienwareAndy New Member

    Joined:
    7 Dec 2009
    Posts:
    3,421
    Likes Received:
    70
    Pics/vids or it didn't happen !
     
  14. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    Really looking forward to seeing how the TS140P performs on your 920!
    Both with stock TY-147 and TY-143s'. Do you have a second TY-147 or TY-140 (same fan) to run push/pull with?

    I think PH-TC14PE & NH-D14 are best reference coolers simply because more good reviews have at least one of them in their cooler list.

    Edit:
    Temps or it didn't happen! :D
     
  15. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    So... now that I've managed to log back into my account, I can share this pic I've been itching to post for 24 hours LOL.

    D14 up and running with two AP-15s on a controller. Early tests show this setup soundly beating a standard TRUE but this is not high TDP, which is where the TRUE really excells. I wonder if I'll be able to reproduce the same results as poparamiro at Lab501, where the TRUE (copper) is only 1-2C behind the D14. We shall see. ;)

    Expecting delivery of my TS140P and TY-143 in about an hour, pics to follow shortly.

    Thanks again to TaRkA for this most generous loan. :thumb:

    [​IMG]
     
  16. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    Inded. First site was down for repairs / upgrades and than had problems with my password. :(

    Could you do me a favor? Monitor the temperature of air going into cooler. This way we know for sure how well cooler is performing.

    For some strange reason most testers use room ambient even when testing in a case. In most cases (no pun) the cooler intake temp is not the same as the room temperature. Even on bench testing it is often different as test runs progress.

    This is easily done using a cheapo digital indoor / outdoor with wired remote sensor and attaching a 20-30cm piece of stiff insulated wire to sensor with an all plastic clothespin on other end. Clip clothespin to case and bend wire so sensor is about 3cm in front of cooler intake fan. ;)

    Mine looks like this:
    [​IMG]
     
  17. sonicgroove

    sonicgroove Radical Atheist

    Joined:
    16 Mar 2011
    Posts:
    2,426
    Likes Received:
    132
    I swear by my Titan Fenrir. one silent 120mm fan that keeps my 2500k @ 4gHz, 2 degrees above ambient on 2 cores, and 4-5 degrees on the other 2 at idle. It never goes above 48 degrees even using Prime95. It's been good enough to take my CPU to a smidge under 5.2gHz without throttling or bluescreening. All this for under £30.

    I actually bought an AIO (H60) to replace it, but it's performing so well still, that I don't have the heart.
     
  18. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    @sonicgroove, most coolers are good enough to do the job. Only a handful of coolers are good enough to do any job, and the TS140P looks set to be one. ;)

    @doyll, already there - my plan is to measure room temperature in three separate locations, one of which is in close proximity to the PC. I agree that ambient temp is a very difficult thing to pin down as there are pockets and currents of warmer/cooler air, so as you say the best way to measure it is either to take one reading from close to the PC or get a mean value from a number of zones in the room.

    In other news, the beast has arrived. I'll let the pics do the talking. The testing is going to take a LONG time so don't hold your breath, but I do hope to be done by the end of the week.

    [​IMG]
    [​IMG]
    [​IMG]
     
  19. doyll

    doyll Member

    Joined:
    1 Jul 2011
    Posts:
    726
    Likes Received:
    15
    Love the pics! Especially the TRUE copper. :D

    TRUE Spirit 140 Power looks bad with the TY-143 fans!!

    Are you going to test NH-D14 with TY-143 fans too? Maybe do load tests at 1000rpm, 1100rpm, 1200rpm, 1300rpm, 1500rpm, 1800rom, and full speed (the maximum speeds standard cooler fans run at). Than use them as reference fans in all cooler tests.?

    I really think you need to monitor the temp of air going into cooler. Room ambient is really immaterial.

    Think of the room as a house with central heating, computer case as lounge with CPU and GPU being a fireplace and a space heater. The room will be the same temperature as the house.. But if you light up the fireplace and turn on the space heater it will become much warmer than the rest of the house. :D

    Sorry, it's your deal and your decision. I'm just an observer and thankful you are doing some of this testing.

    PS: should have a TS140P in a week or two. ;) As we both have very similar systems maybe we can get a reference cooler reading on both systems. Than we can combine test results giving us both a larger database of coolers for comparisons.
     
  20. LennyRhys

    LennyRhys Oink!

    Joined:
    16 May 2011
    Posts:
    5,822
    Likes Received:
    313
    No problem - I get what you mean about intake temp and I did some quick readings. I get the same temp for intake as I get for room ambient, but it is a pretty cold room with no dedicated heating (just my PC and big-ass NEC monitor which is like a radiator, lol). I'm mindful that the intake temp may rise a little during testing seeing that the ram and chipset will all be throwing out heat, so I'm going to set up a small fan blowing on the chipset heatsink to keep that under control. It's a good idea to monitor intake temp though so I'll do it. :thumb:

    And yes I plan to test all coolers with the TY-143s however there will be issues with the D14 because the memory is right under the front fin stack...not even a 120mm fan goes all the way down which is a little frustrating. Memory clearance/interference is probably the biggest drawback with 140mm coolers, and the D14 is not as big as others.

    Since I have already mounted the D14 I'll be testing it first - this should save time and give me a good idea of what to expect from the other contenders. I'll hopefully start tonight.
     

Share This Page