• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

More Radiators the better???

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

narcotis

Member
Joined
Jan 5, 2008
Location
Wisconsin
Is there any reason 3 radiators will be wors than 1?

and im not talking about 3 cheap ones vs. 1 nice one,

i am thinking abought one dual or triple size Danger den and 2 Single size radiators. let me try and explain my loop.....

Cool water goes through the pump to the triple radiator..
icy coald water from the triple radiato goes to the Q6600, exits hot....
hot water goes to the single radiator
, exits cool........
cool water goes to the 2 8800GT,
exits hot.......
Hot water goes th the 2nd single radiator,
exits cool.....
cool water passes through the pump, to the triple radiator.....

Exits icy coald anain, and repeats.....

How does this sound to you all???? it may be over kill but will make sure the water is icy coald when it hits the Heat pumping monster quad :beer:

What are your thoughts???
 
I'd think the added resistence would adversely effect the flow. I've been running a swiftech 2x120mm and for awhile and I tried a single between the cpu and gpu and it did absolutely nothing for temps... possitve or negative.
 
hmm..... i would think it would drop the temps a little bit, how much would that effect the flow?
enough to make it not worth the performance loss,
Should i go with 2 single radiators, or a triple radiator???
and other suggestions?
 
hmm..... i would think it would drop the temps a little bit, how much would that effect the flow?
enough to make it not worth the performance loss,
Should i go with 2 single radiators, or a triple radiator???
and other suggestions?

Maybe two pumps to keep up the flow? I honestly can't say. I'd wait for one of the guys with the giant square cube cases and 3 or 4 radiators to comment.

All I know is that for me (X2 4400, 8800GT) it had no effect, good or bad.
 
well youd need a good pump to deal with the extra restriction, and as far as i know you wont get any more performance from using a extra rad if the rad you have now can deal with the heat load fine
 
well youd need a good pump to deal with the extra restriction, and as far as i know you wont get any more performance from using a extra rad if the rad you have now can deal with the heat load fine
Yeah, won't a 3x120 handle like 300 watts of heat effeciently or something nuts like that?
 
A radiator will not make icy cold water, the only thing a radiator can do is lower to ambient+


and no...

from one block ot another there is only a 0.5-2c difference.
 
Swiftech MCR220-QP Dual 120mm fans ------------------------ 320w

So I guess that explains why I saw no difference. What does the average CPU/GPU combo produce as far as heat in watts?
 
You have a basic misunderstanding of how watercooling works. There is normally only a temp difference of about .5c or less across an average loop. Water carries a small amount of heat with it that when combined with good flow and rad, results in heat dissipation into the surrounding air.

When it comes to rads, there is a balance between flow, surface area, fans, and restriction. It is better to run larger rads as this minimizes the extra tubing restriction that having multiple rads causes. You also have to look at having a pump that will keep up with the blocks/rads etc. that you throw at it. That being said, if your pump can keep up a good flow rate (1.5gpm) or so, then having more rad surface is going to result in better cooling up to a point, depending on the heatload. If one triple rad can cool 320w, but your heatload is only 220w, then adding a second triple rad isn't going to do a thing for your temps or overclock.

You would be better off going with one triple rad than three single rads.
 
With three large rads he could get away with passive and have no fans at all aka a perfect HTPC...

If 2 rads already gets the water to ambient a third rad will not make any difference the water will still be ambient temp.
 
There are two factors to consider with water cooling, Flow rate, and heat disipation from the radiators.

If you can maintain a good flow rate 1.5gmp+ then the usage of large radiators, or even multipel radiators MAY benefit you, depending on what you are trying to do. Running 2 120.3 radiators with the fans oin high is pretty much pointless. But if you can maintain good flow, then you could run two 120.3 radiators with the fans on VERY low speeds for virtually silent cooling.

It all a matter of what you are tryin gto acheive. If you are pumping TONS of heat into a loop, then you MIGHT benefit from a multiple radiators. Or if you want to go with near silent, or even passive cooling.
 
With three large rads he could get away with passive and have no fans at all aka a perfect HTPC...

If 2 rads already gets the water to ambient a third rad will not make any difference the water will still be ambient temp.

Well it would be perfect except that the HTPC would end up being larger than most people's televisions/cabinets/TV related furniture.
 
With a decent pump the rads can be remote - like in a cabinet base or something. No reason to mount them inside the case, especially a smaller case.


My philosophy is "More is Better!"
As long as the temps keep going down (or temps stay the same and noise levels go down) you're good to go ... :)
 
really??

didnt think that pump packed enough power to pump to some rads that are a far away from the main comp/htpc
There's about 11' of 1/2" tubing (round trip) from my computer case to the rads, ~15' for the whole loop. A single MCP655 can run my loop (see sig) on P3 with very little change in CPU temps from both pumps running @ P3 or even P5. I tested it many times when I first set up my loop. I don't know how well this will format on the Net but these were the results of several 10 min. tests using CoreTemp at 1 sec epochs:

Pumps - load ........... All data(StdDev) ...... LastMin(StdDev) .. (Avg)All/LastMin

2805 (255x11) @ 1.53v
off/P3 - 2xPrime95 37.3/52.5 (.703/1.26) 37.1/52.2 (.602/.647) 44.9/44.6
P3/P3 - 2xPrime95 36.4/51.6 (.765/1.31) 36.6/50.7 (.621/1.04) 44.1/44.2
P5/P5 - 2xPrime95 36.3/51.8 (.790/1.52) 35.8/51.3 (.593/.946) 44.0/43.6


2706 (246x11) @ 1.45v
off/P3 - 2xPrime95 32.5/47.0 (.777/1.03) 33.4/47.8 (.610/1.14) 39.8/40.6
P3/P3 - 2xPrime95 32.0/47.1 (.783/1.18) 31.6/46.5 (.585/.567) 39.6/39.0
P5/P5 - 2xPrime95 32.2/46.6 (.737/.996) 32.3/46.0 (.733/567) 39.4/39.2
P3/P3 - idle ............................................. 17.4/26.5 (.993/.977) ----/22.0


This rig runs 24/7 @ 2706 but I turn it up to 2805 once or twice a year (2-4 weeks at a time) for SETI races ... :)


(Excel was used to generate averages and standard deviations from the CoreTemp logfile)
 
Last edited:
Back