DuckieHo

On every single water pump, flow rate and pump head are the two most important specifications. Flow rate is how much water can be moved in time period without any restriction. This value is usually given in gallons per hour (GPH), liters per hour (LPH). A benchmarking procedure would be simply to see how much water can be moved from one tank to another in a hour. Head is how high or hard water can be pushed at full load. This value is usually given in feet or meters. A benchmarking test would be to see high the pump can push water in a thin vertical tube.

All pumps list their pump rate but not all pumps list their head. Why? Flow rate is usually a nice big number... but is an absolutely useless value without a head value. You will never use a pump in a watercooling loop without any restriction. Waterblocks and tubing bends all create restriction. For example, a pump lists its pump rate at 1000GPH but has only 1ft of head. Any type of restriction would seriously drop the actually flow rate. If there is enough restriction, the pump may not actually move any water. While the pump can push the water very fast, it simply cannot push the water hard enough.



Here is another example of an actual pump. The solid black line is for the Danger Den - DDC-12V REV2 "PLUS" while the blue line is for an imaginary pump. The DDC-12v has a flow rate of 425 LPH and a head of 3.5m. In a empty tank without restriction, it can move 425 LPH. On the other hand, it can lift a column of water 3.5m vertically but it will not be moving any water since that is full restriction. Real world performance would obviously full some where between these two points. You want your water to be moving but the loop will have restriction. The imaginary pump has a has a flow rate of 425 LPH but a head of 1m. Let say is would require .5m of head overcome all restriction and move water. The DDC-12V would still be moving water at 390 LPH but the other pump would be doing only 275 LPH. If there was .99m of restriction, the DDC-12V would be at 350 LPH and the other pump would be at near 0 LPH. Be sure to check for head and don't be fooled by high flow pumps!

Attached Thumbnails

 

__________________
To answer most of your questions: (1) a fridge cannot cool a PC (2) 64-bit OS for over 3GB or so (3) PCIe 2.0 is backwards compatible with PCIe 1.x (4) Resolution, not screen size (5) If you have a question, it is not news (6) Read TOS (7) Report, not respond to Spam (8) Uninstall nTune (9) Single/Non-Modular Rail PSUs are NOT better than Multi-Rail/Modular (10) Edward is the Law!

ira-k

A easy way to think of "Head" is to think of it as pressure....Which it's not,but the higher the head the more easily a pump can can push more water through the loop....Most good WC pumps have at least 9' of "Head" and up...

__________________
**Water Cooling Essential Threads** Overclock.net's Downloads Section** TEC Guide:Basic Principles of TEC's

**E8600 @ 5.3Ghz 10x **E7200 @ 4.5Ghz 9x **5000BE @ 3.5Ghz 10x**150 Opty @ 3.2Ghz 12x**

mxrider450

I think Flow rate is overrated, i only get about 180 gph but i have a head rate of 22' and a max 14 psi. But i have no problems.

d3daiM

Some excellent info.

rep +

__________________

Performance Shroud FAQ | Extensive Guide to RMAing

DuckieHo

180GPH = 681LPH
22' = 6.7m

You get only more head and flow rate than the D5 or DDC-12V. What pump do you use?

__________________
To answer most of your questions: (1) a fridge cannot cool a PC (2) 64-bit OS for over 3GB or so (3) PCIe 2.0 is backwards compatible with PCIe 1.x (4) Resolution, not screen size (5) If you have a question, it is not news (6) Read TOS (7) Report, not respond to Spam (8) Uninstall nTune (9) Single/Non-Modular Rail PSUs are NOT better than Multi-Rail/Modular (10) Edward is the Law!

ira-k

Sounds like an Iwaki WMD/MD 20 RLZ.....Good pumps...

__________________
**Water Cooling Essential Threads** Overclock.net's Downloads Section** TEC Guide:Basic Principles of TEC's

**E8600 @ 5.3Ghz 10x **E7200 @ 4.5Ghz 9x **5000BE @ 3.5Ghz 10x**150 Opty @ 3.2Ghz 12x**

Mastacator

Nice info Duckie!

Its always more worthwhile to have more "feet of head" in reference to a pumps capabilities. Flow rate will follow if you've got good head pressure.

Feet of Head is a measurement of pressure though.....

It takes 0.433psi to lift a column of water 1 foot. That means 3 feet of head = 1.299psi

Going back to the pump flow curve Duckie posted.....
Since Gpm is usually used, here in the US, for hydraulic calculations, it will be used, as that is what I use everyday.
The pump flows 425Lph (Liters per hour), which is equals 1.87Gpm (gallons per minute), at 0 feet of head, or 0 pressure. Well, we will never have 0 pressure, right. As Duckie said, know the flow at 0 head does us little good.
Trace the curve back further so something reasonable....
265Lph @ 2m of head. OR
1.16 Gpm @ 6.56 Feet of Head

Remember above.....? (0.433Psi to lift a 1 foot column of water)

So, a 1 foot column of water exerts a pressure at the base of 0.433 Psi, therefore..

The pump is capable of a flow of 1.16Gpm @ 2.84Psi
Considering the size of an average watercooling loop...1.16Gpm is pretty substantial!

Of course the friction loss with the given waterblock(s), radiator(s), fittings, and tubing will take away from the Psi its actually flowing at, so you may be getting a flow of 1Gpm at a diminished pressure, say 1Psi.
The key though is knowing the capabilities of the pump, if it has a good feet(meter) of head, you know it will be able to handle the restrictions put into the loop and still maintain a good flow rate.

Note: The pump curve posted also shows power consumption... Notice the power consumption, in Watts, goes up with the Head pressure. The pump has to work harder as more pressure is asked/required of it.

The feet of head rating on a pump is key, to know if it will handle your waterblocks. A higher head pump would be necessary to make a Storm waterblock perform well, due to the restriction and pressure requirement of the block. An Apogee block is less restrictive and wouldn't "require" a higher head pump, but certainly isn't a bad idea to have one.

ira-k

It's good to know that about the PSI...Thats the way I thought of it but I thought the were a little different....I have a 22' head pump with just a CPU block in it and my pressure gauge on it is reading 5 -Psi, so that means I'm using 18ft of my head for the loop? Ouch....EDIT: Mastercator I just went and looked...This test is showing high Psi left at max head...And I know I've seen that in other tests also...Thats why I always thought of head and Psi as different...[H]ard|Forum - DDC Top & Pump Performance Comparison (Petra's, Radiical, Alphacool, etc.)

__________________
**Water Cooling Essential Threads** Overclock.net's Downloads Section** TEC Guide:Basic Principles of TEC's

**E8600 @ 5.3Ghz 10x **E7200 @ 4.5Ghz 9x **5000BE @ 3.5Ghz 10x**150 Opty @ 3.2Ghz 12x**

Mastacator

You've got the stinger with the "jetted barb" inlet on it right? Thats why your pressure has been eaten up, that jet makes alot of backpressure/friction. Your guage may not be completely accurate either. Good thing you've got that much head from the pump to start with though!!

I'm not sure what you mean by: "This test is showing high Psi left at max head..."
I looked at the review/test..... just unclear what/where your noticing there.

ira-k

Thats probably what it is, plus I left my tubing a little over long on my rad also....The first test there for max head....Am I reading it wrong or what? Isn't that showing the Psi it takes to get to max head? Thanks EDIT: Never mind I see now....Thats just his measurement of the Psi at max head....I was looking at it wrong...

__________________
**Water Cooling Essential Threads** Overclock.net's Downloads Section** TEC Guide:Basic Principles of TEC's

**E8600 @ 5.3Ghz 10x **E7200 @ 4.5Ghz 9x **5000BE @ 3.5Ghz 10x**150 Opty @ 3.2Ghz 12x**