1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Rads flow and cooling

Discussion in 'Overclocking & Cooling' started by freeboy, Sep 9, 2013.

  1. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    There is not much info outside of Martins testing to find ino so I'm asking here.

    For EXTREME cooling where one cares only about cooling and no regard to noise are there any testing with rads compared at ultra high speed airflow?
    I find the reviews limited when they cut off at 2500 pr 2800 rpm.
    Not that they do not have a wealth of info, I'm just wondering about some rads that may perform better, or worse with higher airflow.

    So, are there any testing of rads with extreme high flow? Thanks

    My subject fans
    Black ice gtx 360
    UT 60 and Monsta
    Swiftep Xp mco320
    Ek coolstreem ? Forgot the exact name. They price in tripple and from 73$ to 120$ and I'm thinking the thick Monsta and. Black ice will perform well with very high flow.
    Any reviews hiding out there other than Martinsliquid?
     
  2. Feänor

    Feänor New Member

    Joined:
    Oct 4, 2005
    Messages:
    512 (0.16/day)
    Thanks Received:
    238
    If you don't have enough contact surface to transfer heat from the rad to air, how high is your airflow won't matter at all, you'll have reached the max cooling that particular rad can accomplish. Getting 0,1 degree better temp going 4k rpm over 2500 is pointless.

    It is way better to get thick radiator and high static pressure (pushing air through a thick rad is way harder than a slim one) fans than a slim one with high speed fans.

    There's not much use getting a 5000 rpm fan (and harming your ears) when a 120x38mm or thicker fan can do the same/better job, while producing higher static pressure and way less noise!

    In general, the problem is fitting the thick rad +thick fan in a(ny) case: not many can fit this combination.

    As far as i know, there's no review of 3000+ rpm fans, probably because no one can game with three of them running full speed beside himself...
     
  3. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    Ok
    I'm not talking fans
    I'm talking about testing the bigger rads with higher pressure... Obviously the thicker rads ala Monsta or more restrictive should climb away from the smaller swiftech etc.
     
  4. Feänor

    Feänor New Member

    Joined:
    Oct 4, 2005
    Messages:
    512 (0.16/day)
    Thanks Received:
    238
    Then it all boils down to how much surface the rad has. Obviously, the thicker the better, assuming the length, width and fin density (in fins per inches, FPI) are equal.
     
    freeboy says thanks.
  5. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    I was re reading martins review of the hl black ice gtx 360 where he mentioned the rad might need more air than the fans where giving, paraphrased....other reviews likewise use lower pressure fans.
     
  6. MxPhenom 216

    MxPhenom 216 Corsair Fanboy

    Joined:
    Aug 31, 2010
    Messages:
    9,959 (6.74/day)
    Thanks Received:
    2,221
    Location:
    Seattle, WA
    Watercooling and "EXTREME COOLING" should not be used in the same sentence.
     
  7. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    That's true
    We are talking the going to the edge and pushing it....
    Extreme water cooling
    Lol
     
  8. MxPhenom 216

    MxPhenom 216 Corsair Fanboy

    Joined:
    Aug 31, 2010
    Messages:
    9,959 (6.74/day)
    Thanks Received:
    2,221
    Location:
    Seattle, WA
    Well you cant get temps better than ambients, unless you use a water chiller.
     
  9. m1dg3t

    m1dg3t

    Joined:
    May 22, 2010
    Messages:
    2,247 (1.42/day)
    Thanks Received:
    513
    Location:
    Canada
    Think Stren has done some testing similar to what youre asking... IIRC there is a point of diminishing returns

    MCR320 is still best bang for buck.
     
  10. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    thanks.. its not about bang for bucks but can I get another little bit of temp to help my OC numbers! lol

    So, the latest additional rad did not give the improvements I was hoping it would... BUT the change of the old Danger Den block to a Koolance 380i gave big numbers... over 10 c! also added a newer better pump but honestly i was and am pumping a good amount of water...

    So I am now at 720... 2X360 with an SR 1 and a GTX.... I am getting 3 more deltas soon and will do push pull, but honestly I dont think the rad # 2 will get any more performance, Im sure its maxed out now... so Ill probably add the old pump and another rad.... then cray about how miserable is is to be at ambient and still not overclock to the moon and then Ill take rad three and drop in something cold...


    BUT
    I did IBT with rad 2 fans maxed.. 3 deltas afb1212GHE... OMG these guys rate high for pressure... and they are shrouded... I dropped at 4.9 OC from prior numbers tripping the 99 100 barrier to low 80's!!! very nice.... so the rad and the new block have been helpfull.. I still think its too close to 100 to get 5.0 or 5.1 stable ... but I had 5.0 validated and not stable before so maybe some voltage tweeking...
    Anyway for all who commented THANK YOU!!! Its been crazy as my old coolant wasn't entirely gone and caused foam, and taking the system apart to clean was a nightmare!!!

    So. Im using the heat ir gun to tell temps... its interesting to see the differentparts of the mobo and the varies temps...

    So to summarize the not high OC testing with 1st rad... 2nd rad and both rads show minimal increases in performance.. the high oc shows dramatic changes... but both rads together again allot better than the GTX with the deltas.. those guys are monsters ...

    lets see 4.5 Oc was high 50's ... 4.9 burn test low 80's...
     
  11. erocker

    erocker Super Moderator Staff Member

    Joined:
    Jul 19, 2006
    Messages:
    39,646 (13.30/day)
    Thanks Received:
    14,037
    You should really think about de-lidding your CPU. You just aren't going to get the temps you want otherwise. The transfer of heat from the CPU to the IHS is the weak link.
     
    MxPhenom 216 says thanks.
  12. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    yep, I hear you, I so wanted "the infinite rad addition" to be my silver bullet, unfortunately even running ambient water with the top tested cpublock water isn't getting it done..
    OK
    so.. should I not fear the hammer and vice and just get on with it???? Thats sure is a small chip!
     
  13. erocker

    erocker Super Moderator Staff Member

    Joined:
    Jul 19, 2006
    Messages:
    39,646 (13.30/day)
    Thanks Received:
    14,037
    Just be careful! For me, it's not worth doing for a few hundred Mhz.
     
  14. gdubc

    gdubc

    Joined:
    Aug 14, 2013
    Messages:
    37 (0.09/day)
    Thanks Received:
    5
    Location:
    Loveland Colorado
    I would like to see a test like you mention but on a rad like the aquacomputers. They seem to be considered a poor performer but I wonder what they would be like with something super high speed/pressure. Otherwise, I think most rads are built geared toward use with a certain type of fan over another.
     
  15. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    Did u see my comments on results? Only a few C from huge delta/rad But major gains from switch to CPU block... I do not think pump change made much as the old one pumped great and had ultra low restriction loop. The new loop is much more restrictive.
    I'm using a high quality or gun for temps... Ones anyone know of an inline probe setup that's not so expensive? My friends at frozen CPU have 100$ setups, down to 50$....
    Anyone have thoughts on an inexpensive ware temp tester?
     
  16. mm67

    Joined:
    Jul 11, 2010
    Messages:
    140 (0.09/day)
    Thanks Received:
    47
    Location:
    Finland
  17. fullinfusion

    fullinfusion 1.21 Gigawatts

    Joined:
    Jan 11, 2008
    Messages:
    8,226 (3.37/day)
    Thanks Received:
    2,018
    You know what I do? Keep it simple. I have an old cooIT tec cooler. I stripped the water block off the unit and just plumbed it into the res and iit just circulates within the res and chills the water.. I.ve had condensation built up on the res but just one time on a hot and humid day. It was fun and gave great results but the power required to run it out weighs the benefits of running 24/7.

    Move north is another option haha..I actually froze my loop one night from leaving the window open all night in -34c temps. I have pictures on tpu on some thread a year or so ago..
     
  18. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    Im trying to stay with only liquid..
    otherwisee I probably would do dry ice..
    So, I just practice delidding on an old duo chip.. with razor.. seems straight forward..
     
  19. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    Is there any reason not to use highly rated tuniq TX-4 if I de-lidd?
     
  20. m1dg3t

    m1dg3t

    Joined:
    May 22, 2010
    Messages:
    2,247 (1.42/day)
    Thanks Received:
    513
    Location:
    Canada
    Because CoolLaboratory Liquid Pro/Ultra is much better. Delidding isn't really worth it if not using CLP/CLU
     
  21. fullinfusion

    fullinfusion 1.21 Gigawatts

    Joined:
    Jan 11, 2008
    Messages:
    8,226 (3.37/day)
    Thanks Received:
    2,018
    So using that pro liquid metal stuff... Say I de-lid how to I keep the top on besides just covering the die and clamping it down. Does that paste cure and after time more less weld it self to the lid?
     
  22. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    I read reviews saying that tx-4 was BETTER ? please quote your sorces if saying CLP is preferred and tested better
    thanks
    I know its popular...
     
  23. freeboy

    Joined:
    Feb 17, 2007
    Messages:
    979 (0.35/day)
    Thanks Received:
    26
    ok.. I de lidded and used tx-4...
    I saw immediately 12 C drop in temps as well as a more level distribution between the cores... yippee.. Talk about a stressor though cutting off the lid with a razor blade.. I just could not bring myself to striking my defenseless chip with a hammer! lol

    So.. I may test and redo the paste, but Ive found a little paste goes a long way...
    Fyi, my industrial work blades where too thick, so I used an oldscholl shavers razorblade... the one time it slipped I was sure I had gone way over the die and screwed myself.. just very relieved its done and tests good..
    Pre block and chip delidd almost thirty C drop.. but testing just the new de lidd dropped up to 16C and average of 12C.. nice
     
  24. m1dg3t

    m1dg3t

    Joined:
    May 22, 2010
    Messages:
    2,247 (1.42/day)
    Thanks Received:
    513
    Location:
    Canada
    Yep, don't know yet if it went dry :eek: I have heard it does dry out with time, so may be best to swap it out every few months? My system is a pig so it'll be fun when i eventually pull the heatspreader back off :laugh:

    I just hold the spreader and clamp it down, you can use silicone or some type of glue for a more permanent solution if you choose :)

    I'm only using this between the die/spreader, the spreader/WB gets STD grease... Whatever i have on hand, be it AS5 or NHt1 or similar...

    Seriously? LoL

    How about you quote your sources stating that Tx4 is better. You are the only person i have seen make that comment.
     
  25. de.das.dude

    de.das.dude Pro Indian Modder

    Joined:
    Jun 13, 2010
    Messages:
    7,652 (4.92/day)
    Thanks Received:
    2,017
    on the contrary.. the lesser the contact area there is available b/w the air/rad interface, the higher the air flow the better the cooling.

    however.. if you look at the previous step, i.e. flow of heat from coolant to the rad, and if the rate of heat transfer of this is less than that of the rad-air, higher airflow will not make any difference.
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page