Cooling A Watercooling Pressure Question

Discussion in 'Hardware' started by Damouse, 7 Mar 2010.

  1. Damouse

    Damouse Thats no moon!

    Joined:
    27 Jan 2007
    Posts:
    211
    Likes Received:
    0
    Hey all.

    Ive got a rather technical question about water cooling. Im in the planning stages of a big scratch build, and this is one of those weird and random threads people post when they hit a design block.

    Lets assume the craziest water cooling setup. Imagine a server, basically, cooled by one singular WC loop. Im looking for problems that would arise from pumping the entire system with one pump; either from top to bottom or from bottom to top.

    Anyone have any anecdotal pictures of too high a pressure blowing out water blocks, or connections? Lets assume im considering pumps that put out 1200 gph with a head of 15ft.
     
  2. Burnout21

    Burnout21 Is the daddy!

    Joined:
    9 Sep 2005
    Posts:
    8,614
    Likes Received:
    197
    You thinking of using a Iwaki RD-30?

    Just seems to work out that way. 15ft of head (not actually a measurement of head by the way. But upon factoring in density of water and gravity you get 0.4485 Bar of pressure)

    To be honest i have never seen a loop get so restricted to see anything blow out. Personally if i was to use a Iwaki RD-30 i would avoid arcylic components like the plague.

    Acrylic might be nice to see inside waterblocks and reservoirs however it fails when pressure and temperture differential is applied to it. I would expect to see cracking after a few months with use of a high pressure loop.

    [​IMG]

    Bloody Americans and their silly units! However a RD-30 at full speed is shifting 17 L/min at a pressure of 0.6 Bar.

    Now if you compare that to 3/4in Tygon which has a maximum pressure rating of 36 PSI at 76 degrees F, ergo 2.5 Bar at 22 degrees C.

    So even if the loop temperature is a little higher than 22 degrees you still have a good safety factor in the tubing .

    So in other words, you should be fine if you avoid thin walled tubing and cheap components.
     
    Last edited: 7 Mar 2010
  3. Damouse

    Damouse Thats no moon!

    Joined:
    27 Jan 2007
    Posts:
    211
    Likes Received:
    0
    I wasnt thinking of a specific pump, i was just taking a good look at aquarium pumps.

    Thanks for the response, its helpful.
     
  4. Frohicky1

    Frohicky1 Awaits his moosey fate . . .

    Joined:
    16 Apr 2009
    Posts:
    1,661
    Likes Received:
    9
    You could think of it this way:

    Force = Mass x Acceleration

    The weight (force) exerted downwards by the water as it travels vertically must be balanced or exceeded by the force provided by the pump. The amount of mass in the vertical bits will increase the larger the loop is (mass = density x volume), so more force is required to balance it.

    It does beg the question though, does it matter how big it is if it's all entirely horizontal? :geek:
     
  5. Burnout21

    Burnout21 Is the daddy!

    Joined:
    9 Sep 2005
    Posts:
    8,614
    Likes Received:
    197
    the main reason to go big power on a pump is if you have a lot of gear within the loop.

    Most people running an RD-30 tend to have multiple radiators such as 120.3's or 120.4 if their lucky to find them, all in one loop for a total over kill.

    Pumps with this sort of power like the RD-30 are total overkill and i can think of only 2 mods on the bit-tech project log forum that could have used one.

    Basic and advanced watercooling would not require much more than a typical Laing DDC-1T Pro, only truly mad builds might require more power, but then all you do is piggy back a pair of DDC-1T's together.
     

Share This Page