Typical Delta T across block?

[Barryng]

I am planning to switch to water cooling so I can eliminate the annoying din of the fans, especially the big Delta on my Swiftech CPU cooler. I want to size the pump properly so I calculated the temperature increase across the block based on a flow rate of 75 gph and a 75 watt heat input from the CPU (worst case). The Delta T across the block calculated to just under 0.5 degree F.

I found this to be suprisingly small but I have no experience with tiny heat exchangers to go by. Typically the discharge temperature of the water should approach the temperature of the heat source. From that assumption I expected a much larger temperature increase across the block.

What kind of temperatures rises are seen in actual installations? Also, what are the flow rates (if known) associated with this data. Any help would be appreciated.

 

[Tiger]

I was getting 0.2C with the BeCooling jagged edge block and a flow rate of 3L/min and reported delta bewteen coolant and CPU was 10C. That's the Duron 800 at [email protected] vcore and generating around 70W

 

[Aesik]

It's not surprsing at all that the delta T is so small. In the case of water blocks, we are moving a relatively large amount of water compared to the energy that is being dissipated. Most heat exchangers for other applications are designed to remove a large amount of heat from the source, but are not necessarily concerned with keeping that source at a minimal temperature. Whereas with water cooling a CPU, the main goal is to keep the temperature source (CPU) as low as possible.

 

[Barryng]

Aesik, very good point! I guess the obvious isn't so obvious sometimes.