Flow to temperature

robmac65

robmac65

robmac65

Associate
Joined
1 Jul 2014
Posts
60
I have just changed my xspc x20 750 pump/res for EK Water Blocks EK-D5 Vario X-RES 140 (incl. pump) but my temps have gone up under stress from max @ 51c to max now 56c.

I have the d5 set to 4 and it seems way more powerful but can a faster flow rate actually make the temps go up?

My reasoning is that as its flowing faster then its not spending as much time in contact, is this possible?

I have not changed it to slower to try as it means taking mu gpu out and i cant be bothered just yet lol.
 
I think i have sussed it out anyway, simple really, the tube res in now mounted above the gpu and because of that the heat its throwing out is increasing the water temp whereas before the bay res was further away and in a bay cage so it wasn't affected as much, god damn,so it looks way better but poorer temps or looks not as good with better temps :(.

Looks like i need another rad to lower the temps again :eek:, looks like i have made the wrong decision, should have gone with a d5 and a monsoon!
 
Technically increased flow rate can raise temps, it's because blocks (CPU/GPU/etc) need a certain amount of flow to reach their optimum point, before it's reached they won't be able to dump heat to water as quick as they could and once it's reached more flow isn't going to do anything. Radiators on the other hand don't really have a drop off but it's reversed, the faster the coolant passes through them the less time it has to shed heat.

It's not a massive difference as coolant temperature throughout the loop will equalize, anyway so the will only be a degree or two difference between the hottest/coldest points (unless the is insufficient flow).
 
ubersonic said:
Technically increased flow rate can raise temps, it's because blocks (CPU/GPU/etc) need a certain amount of flow to reach their optimum point, before it's reached they won't be able to dump heat to water as quick as they could and once it's reached more flow isn't going to do anything. Radiators on the other hand don't really have a drop off but it's reversed, the faster the coolant passes through them the less time it has to shed heat.

It's not a massive difference as coolant temperature throughout the loop will equalize, anyway so the will only be a degree or two difference between the hottest/coldest points (unless the is insufficient flow).
Click to expand...

Your first point says there is an optimum point for a block and beyond which you get no gain. I'm not sure that's true but still doesn't explain temps going up with more flow.

Radiators don't work like you suggest, more flow will always reduce temps, yes a given amount of water will spend less time in the radiator but to get the greatest delta T you want to keep the radiator as hot as possible, therefore more flow is better.
 
Well I lowered the pump from 4 to 2 and the temps have dropped down 6c, so then to be sure i put it back to 4 and it went up again dropped it back to 2 and it dropped again, make of that what you will.
 
Minstadave said:
Radiators don't work like you suggest, more flow will always reduce temps
Click to expand...

Sorry but no, radiator efficiency works best with less flow. The ideal way to set up a loop is actually to run the pumps/blocks in series then the radiators in parallel because this halves the flow-rate through the radiators without effecting the flow rate on the rest of the loop, it's just almost nobody does it this way because it's more effort and looks crap.

Here's some info on it from a very old PC watercooling guide from back in the day:

When it comes to radiators, you always want to run them in parallel. Why? Well, the longer that the water stays in the radiator the more time it has to cool. If it rushes through multiple radiators it’s only going to cool so much. What's more, radiators in series will be more restrictive to the water flow.
Click to expand...
 
The coolant is in contact with both the blocks and the radiators 100% of the time.
The only time flow rate will affect your temperatures is when the rate of flow is lower than the rate at which heat dissipates through the coolant.
 
Again I don't believe that to be true. The faster you move a coolant through a water block the the greater the delta T between the metal and the block, the greater the heat transfer into the coolant. The same goes for the radiotor.
 
That would be true if you had an infinite supply of fresh coolant, albeit limited by the rate at which the block transfers heat to the coolant.

Since it's a closed loop once you reach a certain speed you're just feeding warm coolant back through the radiator. You're limited by the thermal capacity of the coolant and the rate at which the radiators transfer the heat from the coolant.
 
Increasing the flow rate just moves you closer to that theoretical maximum heat transfer rate in the radiator, but you reach a point where increasing pump power just adds more heat than the radiator can dissapate resulting in temperatures going up. In a single pump setup you're not likely to hit that point.
 
You must log in or register to reply here.
Back
Top Bottom