Pre cpu / after cpu temp difference ?

Randall8686 · 18 Dec 2014 at 21:54

What sort of water temp difference should you get before and after the cpu ? Im guessing if the flow is correct it should be fairly minimal ?

hugobosslives · 18 Dec 2014 at 22:42

never measured it but theory would dictate minimal difference. if flow is good enough it should only have a big change when you first start the high load program. it should even out and maintain a temperature through the setup.

hence why loop order doesn't matter apart from pump/res

Randall8686 · 18 Dec 2014 at 22:54

This is what i was thinking. But i was just about to order a inline water temp probe, wondered if its worth getting two to see the difference.

Disco_P · 18 Dec 2014 at 23:10

You could get 2 if you are really curious but it wouldn't change much.

For the most part people measure water temp to know that it is not too high as such to cause issues. So you would want to put it where water temps would be highest in theory.

Which I haven't...

Randall8686 · 19 Dec 2014 at 06:58

I assumed that a high enough flow would cause very little difference, however there has to be a difference, just wondered by how much.

The aslower the flow the higher the difference i guess.

almighty15 · 19 Dec 2014 at 07:02

You'll only be looking at a 4-6c increase between the water entering the CPU block and leaving. And that increase is when running Intel Burn Test, gaming the temp increase barely breaks 1c.

The water doesn't actually go up in temp as much as people think it does from CPU's, GPU's on the other hand heat it up a bit more.

But it ultimately depends on loop order and radiator power too.

Disco_P · 19 Dec 2014 at 20:03

CPUs also tend to run a lot hotter than the water flowing through the block too.

I assume this lack of heat flow is due to the IHS but I could be wrong.

pastymuncher · 20 Dec 2014 at 14:46

I have inline temp sensors on the inlet and outlet of my case and I see a 0.5-1 degree difference at idle and 1.5-3 degree difference when gaming (depends on the game).

Disco_P · 20 Dec 2014 at 18:26

I assume that in your build all the heat generating components are on the inside and the radiators are on the outside?

Mick-1965 · 23 Dec 2014 at 17:09

Randall8686 said:
I assumed that a high enough flow would cause very little difference, however there has to be a difference, just wondered by how much.

The aslower the flow the higher the difference i guess.

IIRC
It takes roughly 4 watts for 1 second to raise 1ml of water by 1 degree C

An average flow rate is ~1gpm ( ~75 ml per second )

work it out

pastymuncher · 23 Dec 2014 at 19:05

Disco_P said:
I assume that in your build all the heat generating components are on the inside and the radiators are on the outside?

Yes. Both rads are mounted in a box on a windowsill for stupidly low temps.

Jokester · 23 Dec 2014 at 19:10

For normal flowrates it's about 1degC per 250W.

Randall8686 · 24 Dec 2014 at 07:05

Jokester said:
For normal flowrates it's about 1degC per 250W.

Cheers all for your help, going to start the build today, impressed with how non restrictive the rads are, i was thinking my loop should go

Res - rad - cpu - gpu - rad - res ?

I have two rads and a giant res not that the res size should make a huge difference ?

Also my blocks (all ek) came with their own thermal paste but i have some of that fancy diamond stuff from ocuk. Best to use that diamond stuff ? Super impressed with that stuff. done my mums i5 laptop with it and 11c difference