~40c temp difference coolant-cpu

Angst

Angst

Angst

Soldato
Joined
26 Nov 2008
Posts
4,025
Location
Leeds
When my system is on full load stress testing there is about a 40c difference between the coolant and the cpu's core temperature. The coolant runs about 21-24 and the cpu runs at about 60-65. Is this kind of a gap normal or is the cpu block not transferring heat as efficiently as it should.
 
What a strange question.

Your coolant will be close to ambient air temps if the radiator is doing its job and not overloaded.

Your CPU temp depends on the cooling yes but also what you're doing to it.
 
I've no idea what my water temperature is so can't be terribly helpful. CPU under full load and graphics cards idling I'm seeing about 30 degrees between cpu and graphics card, and am convinced that my cpu block is underperforming.

Seems a sensible question to me. Processor to coolant delta is as good a judge of cpu block performance as any I can think of, in the interests of working out if the block is filled with dye, mounted badly, or needs more flow rate it's about the best measure going.
 
JonJ678 said:
I've no idea what my water temperature is so can't be terribly helpful. CPU under full load and graphics cards idling I'm seeing about 30 degrees between cpu and graphics card, and am convinced that my cpu block is underperforming.

Seems a sensible question to me. Processor to coolant delta is as good a judge of cpu block performance as any I can think of, in the interests of working out if the block is filled with dye, mounted badly, or needs more flow rate it's about the best measure going.
Click to expand...

Thanks, im just using the standard coolit block and just wanted to know if it was preforming within the norm because i am currently deciding whether or not its worth it to get an aftermarket block.
 
What's the pump like? I'm confident my ek supreme will perform far better once the flowrate through it is higher. Which coolit is this?

How confident are you in the accuracy of the reported coolant temperature?

From google, water to processor temperature differences of a few blocks. Testing method looked sound except that he didn't bother to use a backplate.
2e50e9k.gif
 
Last edited:
JonJ678 said:
What's the pump like? I'm confident my ek supreme will perform far better once the flowrate through it is higher. Which coolit is this?

How confident are you in the accuracy of the reported coolant temperature?

From google, water to processor temperature differences of a few blocks. Testing method looked sound except that he didn't bother to use a backplate.
2e50e9k.gif
Click to expand...

I dont know what the block is, its just the standard one that comes with the freezone elite and the pump im using is this one.http://www.overclockers.co.uk/showproduct.php?prodid=WC-045-XS&groupid=962&catid=1522&subcat=

So by the looks of that chart the waterblock is preforming pretty poorly then.
 
Depends, if that's their tec thing then which side is giving temperature measurements, the hot side of the peltiers or the side the processor is on?
 
In that case then yes, it appears the block is rubbish. The pump isn't anything special so you want a high flow one, probably the cheap heatkiller or the "ek supreme lt".
 
i would personally get a normal EK supreme and remove the impingement plate

ots coolit, so its not terribly brilliant - are you using standard watercoling hardware otherwise?

got any pics of your loop mate, im rather interested :)
 
rjkoneill said:
i would personally get a normal EK supreme and remove the impingement plate

ots coolit, so its not terribly brilliant - are you using standard watercoling hardware otherwise?

got any pics of your loop mate, im rather interested :)
Click to expand...

Yea apart from the freezone its just an rs120 after the cpu block and i replaced
the standard pump with the faster xspc one.



 
You must log in or register to reply here.
Back
Top Bottom