Cooling 2 4890's

[aspire]

speaking of your guide... you ever going to get around to finishing it?

And ya, all that wattage on a single 320 wouldnt yield the best temps. but aspire knows more than i do so listen to him, lol

Got a few months for Winter break so I'll see if I can't hammer some more of it out...

 

[Inked]

Well thanks Aspire and Burninator, I'll look at your guide Aspire.

 

[Antec-User]

Your pump could definitely handle it however all that on a single 320 rad would be pushing it.

I would recommend adding an additional rad when you decide to cool the gpu's. Right now I'm cooling my CPU, GPU, and Mosfets, and running 1x320 and 1x220 rads.

The 320 will be sufficient for a single GPU and the processor... No need for another Rad..

If the OP was so inclined to do a chipset cooler, I'd start a second loop..

CPU and Chipset on a 220 rad...
dual GPUs on the second with a 220 rad..

I'm prolly going to be picking up a second 4870 and liquid cooling it as well, I need to upgrade to a 320 rad.. I really don't think you'll need another rad for that.

I've found that adding 2 rads to a single loop doesn't get you any better temps than good placement of the single RAD in the first place. I was running dual 220's and my temps went down about 3*c when I ditched the second 220 rad and moved it to the top of the case (Over the 200mm fan).

 

[Burn]

The 320 will be sufficient for a single GPU and the processor... No need for another Rad..

he has 2 4890's.. which is why i said 2 rads would be better.. and judging from his other posts he is upgrading to i7, and those produce quite a bit more heat than PII's..

I've found that adding 2 rads to a single loop doesn't get you any better temps than good placement of the single RAD in the first place. I was running dual 220's and my temps went down about 3*c when I ditched the second 220 rad and moved it to the top of the case (Over the 200mm fan).

unless you just put both in good places to start with... also did you have the fans in push/pull like you do now on both rads when you had the dual 220's?

 

[Antec-User]

^ no I didn't.. And if I switched from dial radiators to one with a push/pull, then the drop in temp changes probably came from the change in fan setups and the placement.

I had both on top for a short time as well with just sucker fans and the temps were the same.

Come to think of it... Dual rads CAN prove two purposes.. Slower moving coolant can absorb more heat, and the RAD can absorb it from the slower moving coolant.. But I did notice a big drop in flow with the dual rads..

 

[Burn]

actually slower flow doesnt mean it absorbs more heat.. at least from what ive read on xtreme systems it doesnt.

so did you have the dual rads stacked? as in one behind the other. rads dont perform at all well like that, unless you have some SERIOUS fans, ike 2500rpm+ i would imagine.

also rads wont inhibit flow too much, adding a rad drops flow by about <.2gpm

 

[Antec-User]

^ oh god no.. No sense at all in stacking rad's IMO.. I had two mounted side by side, in series, on the side of the comp, and then mounted on the top.. Then I switched to the single Swiftech 220 and mounted it right over the top 200 mm fan on the 900 case..

OLD pic.. When I first set it up..

flow was: Pump > RAD > CPU > RAD > GPU > PUMP

Pic after the revision.. Kinda.. Right when I eliminated one RAD and was running PRIME.. Moved rad to the top shortly after.. I'm always tinkering with this computer. I got rid of that crappy display thing.. It wouldn't stop beeping at me..
Flow in this went: PUMP > RAD 1 > RAD 2 > CPU > GPU > PUMP

 

[aspire]

Come to think of it... Dual rads CAN prove two purposes.. Slower moving coolant can absorb more heat, and the RAD can absorb it from the slower moving coolant.. But I did notice a big drop in flow with the dual rads..

If you said this on XS you would be flamed to death.

It has been proven countless times that higher flow rate = higher heat dissipation.

Now, the water might take in a little more heat, but because the total number of times an amount of water goes through the loop, the more times it can take in and then dissipate heat, higher flow rate will always lead to lower temps.

 

[Burn]

so i was right,