Single loop vs dual loop

[lostredneck]

I have never done a custom loop before, always just sticking with AIO's. I'm trying to work up the courage to try it with my new build but there are a lot of things I dont know.

 

I went to the EK loop builder web site and they suggested 2 360mm rads for my loop to cool the CPU and GPU, that part I understand pretty well. What I dont understand is why they suggested putting the whole thing in the same loop. 

 

It seeks to me the water is going to soak the heat from the CPU before it hits the GPU so it will cool the GPU less effectively than it could. Wouldn't it be better to do 2 separate loops 1 for the CPU and 1 for the GPU? 

 

Then I dont understand the dual radiator idea, what's the benefit of running rads in a series like that instead of using separate loops.

 

Thanks

[Pixel5]

there is close to zero difference between dual loop and single loop as long as you have enough radiator surface area and flow in the loop.

 

the water passing through the cooler only heats up by a few degrees at most before it comes out again so if the water going to your CPU for example has 25°C it will have well below 30 probably closer to 28° when it comes out and goes to the GPU.

 

the difference is so marginal that you could maybe get a degree lower temps on the 2nd component if you got for dual loop for the cost of two loops.

 

with the dual rads its a similar story but the other way around, one radiator does not mean the water is completely cooled down when it comes out, it will lose some energy and thus temperature but its all about the surface area of the rad so two rads is nearly the same as having one bigger one and how much you need depends on how much energy you want to dissipate.

[lostredneck]

10 minutes ago, Pixel5 said:

there is close to zero difference between dual loop and single loop as long as you have enough radiator surface area and flow in the loop.

 

the water passing through the cooler only heats up by a few degrees at most before it comes out again so if the water going to your CPU for example has 25°C it will have well below 30 probably closer to 28° when it comes out and goes to the GPU.

 

the difference is so marginal that you could maybe get a degree lower temps on the 2nd component if you got for dual loop for the cost of two loops.

 

with the dual rads its a similar story but the other way around, one radiator does not mean the water is completely cooled down when it comes out, it will lose some energy and thus temperature but its all about the surface area of the rad so two rads is nearly the same as having one bigger one and how much you need depends on how much energy you want to dissipate.

I can understand there is negligible difference throughout the loop as a whole, what I'm having a hard time wrapping my head around is the difference in cooling potential on the GPU. Since the water has already been saturated with at least some heat from the CPU it seems like there would be less cooling potential left in the water when it hits the GPU.

 

Again I could be talking out of my ass on this since the water is moving so fast there's no way it's getting anywhere near its heat threshold while passing over either components regardless of they are in a series or parallel. 

[andrewmp6]

The reason ek said to run 2 rads is that your whole loop a heat ratting in watts,Each rad has a max watts of heat it can cool so you need two rads to cool it.If you had room in your case for a thick rad like 80mm with push pull you can run one rad.The single vs dual loops is pretty much dual is for show and looks and would cool the same in a single loop.Like said with the coolant/water moving at its speed it does take a while to come up to temp and equal out.Its been proven cpu to gpu on the same loop is with in 5 degrees of each other max.

[Pixel5]

you need to get away from the concept of the water being saturated with heat, water is not saturated until its boiling and it takes huge amounts of energy to get to that point.

 

as i explained the water coming from one component will only be a few degrees hotter then when it entered and it will not make a difference for the other components in the loop.

[lostredneck]

11 minutes ago, Pixel5 said:

you need to get away from the concept of the water being saturated with heat, water is not saturated until its boiling and it takes huge amounts of energy to get to that point.

 

as i explained the water coming from one component will only be a few degrees hotter then when it entered and it will not make a difference for the other components in the loop.

Ok thanks that does make sense. 

[ninbura]

I'm running 7900X > GTX 1080 > 6700K > GTX 1080 > Capture card all in succession, one loop, with no radiators in-between. If both PCs have been off and I boot everything at once, record temps right away, and then wait a few minutes for everything to warm up the first GTX 1080 in the loop will raise about 2 degrees at idle while the second GTX 1080 in the loop will raise around 4.

Gap becomes around 4 degrees with every component at load, as in the first GTX 1080 goes from around 25c (idle) to 36c (load) and the second GTX 1080 goes from 27c (idle) to 40c (load). Hard to say if this is even entirely because the water is hitting other components before reaching the second GPU as even like cards typically have slightly different thermal profiles. 

Basically as @Pixel5 is saying, it's really a non-issue.