Results 1 to 10 of 10
  1. #1
    Member lowfat's Avatar
    Join Date
    Nov 2004
    Location
    stoon, sask, can
    Posts
    1,187

    radiators: series vs parallel

    Someone on another forum I frequent wants to put multiple radiators in parallel, would it lower temperatures at all?

    His theory that since the flow rate is halved because of them being in parallel, they should dissipate more heat. Is this true?

    btw mods somehow i posted a thread under the stickies forum, you can erase it. sorry.
    Intel Core 2 Duo E6600.
    Gigabyte GA-P35-DQ6.
    4GB Crucial PC2-8000.
    eVGA 8800GTS 320 Superclocked.
    Cooler Master Stacker 830.
    OCZ GameXStream 700W.
    DTek Fuzion + Weapon 302 Heatercore + AquaXtreme 50Z.
    JVC RX6042, Soundstage Center 5 + Surrounds, Athena A1-B1s.

    3D06-11115 3D05-19593 3D03-37303 3D01-61781

  2. #2
    Member crimedog's Avatar
    Join Date
    Nov 2003
    Location
    South of Boston
    Posts
    3,468
    increased flowrate benifits radiators as well as waterblocks, in that respect the person in question is wrong. however they're not as sensitive as waterblocks to flowrate, there's no reason why more radiators (more surface area) with lower flowrates wouldn't dissipate more heat.

  3. #3
    Member
    Join Date
    Jul 2003
    Location
    New Hampshire, USA.
    Posts
    921
    The point of putting radiators in parallel is to increase flowrate where it counts most - to the waterblocks. Rads in parallel cuts down on restriction/head loss (and actually is less restriction in the loop than one single rad).

    Decreasing the flowrate inside the rad decreases the cooling of the water as it's flowing through it, as said above. However, you're gaining twice the cooling area, and I think that's a fair trade. Series seems ineffective to me because the water will already be cooled as it goes through the second rad, adding needless restriction for negligible gain.

    I would say you should always put two rads in parallel if able. You're making both rads as effective as possible, and allowing waterblocks to perform to their max.

  4. #4
    Senior Something Moto7451's Avatar
    Join Date
    Feb 2004
    Location
    LA, CA
    Posts
    3,972
    Do you have data that shows that two rads in parallel has less resistance than a single rad? That doesn't seem to make any sense to me. Unless there is some sort of physics theorem to explain this I'm thinking that two rads in parallel would have at least as much restriction as a single rad if not more because there are two rads.
    Intel C2D e6400 @ 375MHz x8 = 3Ghz 1.328v, Gigabyte EP43-UD3L, 3GB DDR2, GeForce 9600GT, 73GB 15K SCSI HD/Adaptec 29160 SCSI Card, 750GB SATA Drive, 320GB ATA Drive, HDTV Tuner

    D5 Pump Bonneville w/ AC Heater Core Gemeni Cooling Maze 4 Style Water Block

    Sager 8690: Core i7 940XM Exreme 2.13GHz (3.2GHz Turbo), 4GB DDR3, GeForce GTX 285M 1GB, 15.6" 1080 LCD, 320GB 7200RPM HD

    Difference Between D5 Revisions + flow mod Difference between NPSM & G1/4

  5. #5
    Member crimedog's Avatar
    Join Date
    Nov 2003
    Location
    South of Boston
    Posts
    3,468
    Quote Originally Posted by Moto7451
    Do you have data that shows that two rads in parallel has less resistance than a single rad? That doesn't seem to make any sense to me. Unless there is some sort of physics theorem to explain this I'm thinking that two rads in parallel would have at least as much restriction as a single rad if not more because there are two rads.
    well it's less than two in series

  6. #6
    Member
    Join Date
    Jul 2003
    Location
    New Hampshire, USA.
    Posts
    921
    It is physics. You half the amount of force used to push water through the tight bends and tubing of a radiator. With the decreased friction the water encounters (vs if it were going through one radiator), the water moves more freely and less energy is wasted. Although with added fittings to split and combine the tubing, you probably wouldn't net much. I'll try to find a link.

  7. #7
    Senior Something Moto7451's Avatar
    Join Date
    Feb 2004
    Location
    LA, CA
    Posts
    3,972
    Ah ok. I was under the understanding (perhaps falsely assumed) that pressure & the drop across the rad were proportionate so you'd end up with at least the same amount of pressure drop.

    Cool stuff to know.
    Intel C2D e6400 @ 375MHz x8 = 3Ghz 1.328v, Gigabyte EP43-UD3L, 3GB DDR2, GeForce 9600GT, 73GB 15K SCSI HD/Adaptec 29160 SCSI Card, 750GB SATA Drive, 320GB ATA Drive, HDTV Tuner

    D5 Pump Bonneville w/ AC Heater Core Gemeni Cooling Maze 4 Style Water Block

    Sager 8690: Core i7 940XM Exreme 2.13GHz (3.2GHz Turbo), 4GB DDR3, GeForce GTX 285M 1GB, 15.6" 1080 LCD, 320GB 7200RPM HD

    Difference Between D5 Revisions + flow mod Difference between NPSM & G1/4

  8. #8
    Member
    Join Date
    Jul 2003
    Location
    New Hampshire, USA.
    Posts
    921
    Couldn't find any links, just other hearsay. The thread below is a series vs. parallel discussion about radiators. May want to check it out.

    http://www.ocforums.com/showthread.php?t=236190

    Seems that (theoretically) parallel is the way to go with radiators. But ultimately, you'll have to test your own pump/radiators to find practical results.

  9. #9
    Member
    Join Date
    Jun 2005
    Location
    Vallejo, CA
    Posts
    265
    I run 2 80mm rads in parallel on my server, and it kicks butt. At full load it never goes over 33C and it idles at 29C and the fans are not even running at full speed. Putting rad in parallel increases cooling area and decrease pressure drop.

  10. #10
    Member thorilan's Avatar
    Join Date
    May 2002
    Location
    Japan/Daytona Beach
    Posts
    3,819
    the bet cooling i have seen run parrallel single pass radsthat are wide not long.

    the way the delta T works and xym said it well in this
    Decreasing the flowrate inside the rad decreases the cooling of the water as it's flowing through it, as said above. However, you're gaining twice the cooling area, and I think that's a fair trade. Series seems ineffective to me because the water will already be cooled as it goes through the second rad, adding needless restriction for negligible gain.
    the whole idea is to keep the deltaT as wide as possible for max efficientcy and if you can do short wide rads in parralel without adding more restriction from plumbing you will show a benifit.

    if we could turn todays bix3 sideways so that the flow could only travel through 15cm of tubing but much more tubing ( remember turning the tubes on thier sides instead) you would have same serface area but the coolant keeps the deltaT wider .
    the dificulty lies in designing a rad as such and keeping pressure drop even across all tubes. for this reason rads are not that wide in such small aplications. now with higher pressure its definately doable.

    that is why some people do parrallel rads with series pumps.
    697+ water coolled systems built , 0 leaks

Posting Permissions

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