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

Will a 480 Rad be enough for CPU + 2x 980s in SLI?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
No, a radiator with the fans off will still have a temperature differential with the atmosphere, therefore incurring an airflow through the radiator.
That's simple thermodynamics. Period. It's obvious you know nothing about how heat transfer actually occurs.

I understand the graphs. I also understand that a flatter graph will have a slope closer to zero, when you're saying a steeper graph has a slope closer to zero.

How about this, YOU go do some testing. You're the new kid on the block.
You're ignoring the fact that airflow through a radiator has diminishing returns. Real life doesn't ignore that fact.
Just because you have 4x the airflow through this given radiator doesn't mean you get 4x the cooling! THE GRAPH YOU POSTED SHOWS THIS.

You've got what, four experienced and respected members of the forum saying "you're wrong", but you don't realize it? Sad.

LOL, and still nothing to back it up. Post some evidence. I have already.

You'r saying no and just repeating what I posted now. In an enclosed environment the rad will warm the air to the same temp as the rad itself and cooling is zero. In an open environment the heat of the rad causes warm air to rise and induces an air flow. So what, the 480 rad still needs about 500 RPM to dissipate 200 Watts.

Deminishing returns applys at high RPM'S because radiator restriction (like all restriction) increases with speed. More RPM's are needed for the same airflow. That is what the graph shows.
That has nothing to do with at the effects at the lower end and nothing you have said contradicts the fact that a 120 rad at high speed is equal to a 480 rad at low speed. Its time to give it up because its getting pathetic.



A 480 rad that dissipates 200 watts at zero RPM and 10C delta ... I'll take one of those. :)
 
Last edited:
LOL, and still nothing to back it up. Post some evidence. I have already.

You'r saying no and just repeating what I posted now. In an enclosed environment the rad will warm the air to the same temp as the rad itself and cooling is zero. In an open environment the heat of the rad causes warm air to rise and induces an air flow. So what, the 480 rad still needs about 500 RPM to dissipate 200 Watts.

Deminishing returns applys at high RPM'S because radiator restriction (like all restriction) increases with speed. More RPM's are needed for the same airflow. That is what the graph shows.
That has nothing to do with at the effects at the lower end and nothing you have said contradicts the fact that a 120 rad at high speed is equal to a 480 rad at low speed. Its time to give it up because its getting pathetic.



A 480 rad that dissipates 200 watts at zero RPM and 10C delta ... I'll take one of those. :)

You, sir, are perpetuating a pointless argument, and in someone else's thread, no less.

Bottom line is, NOBODY is going to use a 120mm radiator to cool both a CPU and dual GPUs. If you are that nobody, then by all means, show us pictures and temp logs and prove some of the most experienced members here wrong. Otherwise, I would like to continue learning about watercooling in the real world application.
 
You, sir, are perpetuating a pointless argument, and in someone else's thread, no less.

Bottom line is, NOBODY is going to use a 120mm radiator to cool both a CPU and dual GPUs. If you are that nobody, then by all means, show us pictures and temp logs and prove some of the most experienced members here wrong. Otherwise, I would like to continue learning about watercooling in the real world application.


At the risk of quoting myself. Nobody said they would.

So well done pointing out what has already been said multiple times. If that isn't the definition of a pointless contribution to a thread I don't know what is.

I have a spare UT60 120, UT60 480 and Aquaero, and will be more than happy to set up a test that shows them cooling the same ... as soon as any sort of evidence to refute it is provided. Anything will do ... any dodgy looking half test or test results that comes even close to showing a 120 will never cool the same as a 480.
Til then, maybe provide something before you ask others to.
 
Logic and physics are the proof. You are the one suggesting against those. So, either provide your test results, or kindly remove your hands from your keyboard.

Edit: That was more disrespectful than I had intended. I apologize. I probably shouldn't reply to posts first thing in the morning, particularly the ones where someone is being rude for no reason.

Having said that, let me quote myself in saying that "you, sir, are perpetuating a pointless argument...". That was the point in my post. I feel that it was quite the opposite of a "pointless contribution", in that you still don't seem to get it, so let me be frank with you: The only person on your side of this discussion is you. In this situation, the logical and most effective way to handle yourself is to either provide proof or drop the discussion. You have done neither.

The fact that you attack those who you disagree with suggests that you are either adolescent or not very mature. Either way, you would do much better to show those who have proven time and again that they know what they are talking about some respect.
 
Last edited:
I decided to go with two 480 Black Ice nemesis SR2, the loop will be as follows:

Res -> Pump -> bottom 480 Rad -> 2x980 -> 480 rad top -> CPU -> Res.

the Gpus will be in parallel or series still didn't decided but i guess i will be going parallel.

The fans are Corsair SP120 quite edition, so will the temps going to satisfy? or taking a black ice nemesis GTX 480s instead? but i am looking for quite PC but if the temps are going to be huge deference then i will have to rethink.

Should i follow delta-T with these configuration? just to make sure.

Open for suggestions.
 
Last edited:
Loop order doesn't matter as long as the reservoir is before the pump naturally feeding it but I will say to save a few degrees on the CPU since they are the most sensitive, I would reverse the flow and go top rad to CPU, than GPUs, than bottom rad and back to reservoir/pump. The loop will reach its equilibrium say after 30 mins.
 
Loop order doesn't matter as long as the reservoir is before the pump naturally feeding it but I will say to save a few degrees on the CPU since they are the most sensitive, I would reverse the flow and go top rad to CPU, than GPUs, than bottom rad and back to reservoir/pump. The loop will reach its equilibrium say after 30 mins.

I let the loop go after going to the Gpus it goes to top 480 rad to cool it and then go out to cpu then to res/pump, wont thats save the degrees since i am cooling the loop before going to the cpu?
 
I let the loop go after going to the Gpus it goes to top 480 rad to cool it and then go out to cpu then to res/pump, wont thats save the degrees since i am cooling the loop before going to the cpu?

Nope. The loop will reach a equilibrium, meaning it will neutralize after 30 mins. The only way to lower the temps is with heat surface (rads) and a loop running efficiently at 1.0-1.5 GPM.
 
In a properly radded and flowing loop, temperatures at any given point in the loop will be within 1-2c of each other, yep.
 
But how many degrees will differ if i did my loop? witch is Res -> Pump -> bottom 480 Rad -> 2x980 -> 480 rad top -> CPU -> Res.
 
But how many degrees will differ if i did my loop? witch is Res -> Pump -> bottom 480 Rad -> 2x980 -> 480 rad top -> CPU -> Res.

If you choose to configure your loop as stated, than you'll have too much tubing, the rig will look like a mess with tubing everywhere which also can impede in case air flow. Just make the loop simple as possible.

Again, the loop will reach equilibrium. In any loop configuration, the loop temp will end up the same is what I am saying.
 
If you choose to configure your loop as stated, than you'll have too much tubing, the rig will look like a mess with tubing everywhere which also can impede in case air flow. Just make the loop simple as possible.

Again, the loop will reach equilibrium. In any loop configuration, the loop temp will end up the same is what I am saying.

Actually no,

1 rigid tube from pump to GPUs in parallel, 1 rigid tube from gpus to top rad input on left, 1 rigid tub out from top rad to cpu, out of cpu to res so not a lot of tubings in each other and i am talking about the middle, because this way its more easier to do, but perhaps i can manage to get a soft tube from the back direct from the pump to top rad then cpu then gpus then button rad and to res.
 
Back