pressure drop

breadturbo · 1 Oct 2010 at 23:38

Anybody know how much pressure drop affects watercooling at the high end?

Ive been taking a look at this......
(shameless linkage)

im running the EKsupremeLT, and they look to be off the chart at 3gpm which my pump is.

just wondering how much difference i may see if i went for a higher flowing one like the Fuzion/mc-tdx shown on there.

JonJ678 · 2 Oct 2010 at 00:13

Warning: cant see the pic and drunk

pressure drop is analogous to voltage drop. As such gph means fa unless it's vs resistence. Your pump may well do 3gph through some tubing, but perhaps not through a porous wall. A pump that will do 3gph through an ek supreme is a rare toy, but one worth keeping.

Cant remember the q. The ek supreme is awesome at high pressure, but a bad shout for ultra quiet loops.

My apologies, bit lost but trying to help

breadturbo · 2 Oct 2010 at 00:38

thanks

its not that quiet hehe, its a huge garden pond pump

JonJ678 · 2 Oct 2010 at 01:16

Through the wonders of a good woman and water I'm doing a bit better now.

Flow rate is almost independent of wattage removed by radiators, though higher flow rate does slightly improve their effectiveness. The delta T maintained by a waterblock does depend on flow rate, with higher flow rates leading to lower temperature difference between water and chip.

Pressure drop itself doesn't matter, it's more that a high pressure drop indicates that a given pump will struggle to maintain sufficient flow rate through the block. So really you need to know the flow rate through your loop (measuring it is by far the easiest approach) and compare against a graph of delta T vs flow rate. Then choose a sufficient pump. That is assuming you're trying for lower temperatures anyway.

If a waterblock offers great delta t vs flow rate performance, but has a high pressure drop, it'll be awesome with a powerful pump and crap with a weak one. The EK supreme loves pressure, the xspc edge does brilliantly in low flow rate loops. Adding a second 18W ddc in series took 10 degrees off my cpu temperatures, using a somewhat restrictive loop and an ek supreme. Were I using an edge, I wouldn't expect a second pump to help at all.