Overclock.net › Forums › Cooling › Water Cooling › Water flow vs delta T benchmarks
New Posts  All Forums:Forum Nav:

Water flow vs delta T benchmarks

post #1 of 7
Thread Starter 
I am looking for a benchmark of waterblock performance in different flow speed configurations. A quick calculation gives that even a 57 liters per hour water flow is enough to transfer 1 kW heat to the cooler with 15 degree delta between heater and cooler. This gives me an idea that modern water cooling system use pumps with much greater flow speed to increase heat transfer in water block between processor and water. I have a huge radiator with very poor dynamic resistance. It can easily dissipate up to 5 kW of power preserving the 15 degree delta but no pump can pump water through it with a speed above 100 l/h. Now I need to check the performance of modern water blocks in different flow speed conditions. Ive found this review but it is from 2011 year and author didn't mention the wattage of the cpu used in tests. Can you please give me more recent info on the subject.
I still have an option to make two pump design. One pump will pump water through the cpu waterblock and the other one will pump water through the radiator on much slower speed but I want to avoid using the second pump.
post #2 of 7
Quote:
Originally Posted by unclocker View Post

...but no pump can pump water through it with a speed above 100 l/h.
What? Most pumps will do 1 gal/min (225 l/h) through a loop.

Some "food for thought":
No matter what the flow rate is, the water spends the exact same amount of time in contact with any component in the loop....over an extended period of time. In an hour all the water touches each component for the same length of time... no matter what the flow rate is. If the flow rate is doubled there is less contact time during each of the water's circuits around the loop, but the water comes back twice as fast for another circuit.

When the flow rate changes what changes is the amount of laminar flow ie. more or less turbulence, and that's what causes higher heat transfer efficiencies at higher flow rates. There's a little more to it than that, but that's pretty much what's happening.
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
post #3 of 7
Thread Starter 
Quote:
Originally Posted by billbartuska View Post

What? Most pumps will do 1 gal/min (225 l/h) through a loop.
It has a very long pipe, 20 meters. It is not a parallel pipe design. Dynamic water resistance is huge and its increases with flaw speed dramatically. Most water pumps do ~4-5 mH2O pressure and it will be able to pump less than 100 l/h through this monster.
Quote:
When the flow rate changes what changes is the amount of laminar flow ie. more or less turbulence, and that's what causes higher heat transfer efficiencies at higher flow rates. There's a little more to it than that, but that's pretty much what's happening.

Exactly! And most waterblock producers do use greater flow rates to improve heat transfer from CPU to water. Again, 1 kW can only heat 57 liters of water to 15 degree.
post #4 of 7
Have you looked for data on Martin's?
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
post #5 of 7
Quote:
Originally Posted by unclocker View Post

I am looking for a benchmark of waterblock performance in different flow speed configurations. A quick calculation gives that even a 57 liters per hour water flow is enough to transfer 1 kW heat to the cooler with 15 degree delta between heater and cooler. This gives me an idea that modern water cooling system use pumps with much greater flow speed to increase heat transfer in water block between processor and water. I have a huge radiator with very poor dynamic resistance. It can easily dissipate up to 5 kW of power preserving the 15 degree delta but no pump can pump water through it with a speed above 100 l/h. Now I need to check the performance of modern water blocks in different flow speed conditions. Ive found this review but it is from 2011 year and author didn't mention the wattage of the cpu used in tests. Can you please give me more recent info on the subject.
I still have an option to make two pump design. One pump will pump water through the cpu waterblock and the other one will pump water through the radiator on much slower speed but I want to avoid using the second pump.

Flowrate for heat transfer doesn't matter for liquid cooling in our case. Since your loop is indeed a loop the temperature will be more or less at an equilibrium throughout the entire loop, maybe a couple degrees kelvin max throughout the loop. As has been said turbulence is the big thing for our purposes. Most waterblocks don't actually get a huge benefit from flowrate. Their design in of itself produces turbulence. Theoretically radiators receive the biggest benefit from flowrates, since they are more or less just a bunch of straight tubes.

Tubing distance wont effect your flowrates very much, if at all. Since your loop is indeed a loop any pressure lost from pumping it up against gravity is regained when it goes back down. The only way tubing effects flow rates is the bends and surface tension, which typically don't have very large impact.


Something tells me this isn't a normal liqiud cooling loop. Can you give up specifics on your plan?

-Z
The Laboratory
(11 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-8350 Vishera 8-Core 4.0GHz Asus Crosshair V Formula-Z Radeon R9 290X Generic 1333MHz R.A.M. 
Hard DriveOptical DriveCoolingOS
Wester Digital 5400RPM HP CH20L "The Russian Winter" Windows 8 
KeyboardPowerCase
Steel Series Merc Fatal1ty 1000Watt PSU DIY Rig 
  hide details  
Reply
The Laboratory
(11 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-8350 Vishera 8-Core 4.0GHz Asus Crosshair V Formula-Z Radeon R9 290X Generic 1333MHz R.A.M. 
Hard DriveOptical DriveCoolingOS
Wester Digital 5400RPM HP CH20L "The Russian Winter" Windows 8 
KeyboardPowerCase
Steel Series Merc Fatal1ty 1000Watt PSU DIY Rig 
  hide details  
Reply
post #6 of 7
Thread Starter 
Quote:
Originally Posted by ZytheEKS View Post

Tubing distance wont effect your flowrates very much, if at all. Since your loop is indeed a loop any pressure lost from pumping it up against gravity is regained when it goes back down. The only way tubing effects flow rates is the bends and surface tension, which typically don't have very large impact.

Tube distance do affect flowrate by introducing a huge resistance to the flow at high speed. Here is the calculator to play with http://www.pressure-drop.com/Online-Calculator/
Quote:
Volume flow: 0.300 m³/h
Element of pipe: circular
Diameter of pipe D: 10 mm
Length of pipe L: 20 m
Pressure drop: 0.34 bar

0.34 bar is 3.46 mH2O. The speed of a flow at 300 l/h is already big enough. If we increase the speed up to 600 l/h pressure drop will be 12 m of H2O.
post #7 of 7
+1 ^ ^
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Water flow vs delta T benchmarks