Where to put 2nd Radiator into current loop?

BornEvil

BornEvil

BornEvil

Associate
Joined
18 Oct 2002
Posts
859
Location
London
XSPC ResTop and Laing DCC-1T 18W Pump -> TFC 360 Radiator -> DTEC Fuzion v2 CPU Block -> Swiftec MCW-60 GPU Block

I bought another triple radiator (XSPC RS360) to add to my current loop. but where would it be most effective?

for ease of mounting i was thinking:

XSPC ResTop and Laing DCC-1T 18W Pump -> TFC 360 Radiator -> DTEC Fuzion v2 CPU Block -> Swiftec MCW-60 GPU Block -> XSPC RS360 Radiator.

so i would have rad>pump>rad.

or should i stick in between the water blocks?
 
I run my processor and chipset on a seperate loop from my graphics cards, but if you can't do that, i'd stick it after the cpu block, just before the GPU block - especially if your overclocking your graphics card as well.
 
my graphics card is overclocked but it max loads at 40c. my main priority is the CPU temps and the water temps at entry into the CPU block.

i have currently have a inline sensor straight after the GPU block and that is at 30c. i need that lower.
 
Jokester said:
It doesn't matter where you stick the radiator in the loop from a cooling point of view.
Click to expand...

this is not always true. Especially with muliple radiators, you would definatly see different sectors of the loop that has slightly lower temps.
 
Last edited:
If it was me i'd start a second loop with the second Rad. As far as I know, seperate loops work better than one big loop with multiple rads.
 
Jokester said:
If by slightly, you mean substantially less than 1C?
Click to expand...

depends on flow rate, a fast flow system will be more efficient at getting thermal equilibrium. But it is completly obvious the outlet temp would be definatly lower than inlet temps.

But in my case my pumps is heavily under volted, so the effect of multiple pump order should make a difference.

This is all theory, not installed it yet. And if it only give me a 1c decrease, then it will be a 1c decrease. :)

I will not be adding another loop, a triple rad for just the GPU is just not needed.
 
Flow rate doesn't make that much difference either to be honest, as Cathar's tubing size testing showed. For a typical heat dump of 200-300W for a multi block loop the delta T across a radiator is about 0.2-0.3C.

It's not worth getting hung up on where to put it in the loop. Put it where it minimises bends and tube length.
 
my current TFC Xchanger is rated for about 1000W of heat (dunno exactly) but i want to depense the heat dump over 2 radiators.

say about 150W of heat comes from CPU and 100W from GPU.

I want the 150W (CPU) to be cooled with 1 of the radiator then the 100W (GPU) to be dumped with the other. In effect the maximum water temp (at any given point in the loop) would be lower.

I have not seen 'Cathar's results' but all rad reviews i have read showed much higher delta T.

So I think:

XSPC ResTop and Laing DCC-1T 18W Pump -> TFC 360 Radiator -> DTEC Fuzion v2 CPU Block -> XSPC RS360 Radiator -> Swiftec MCW-60 GPU Block.

just so happens the above gives me minimum tube bending aswell :)
 
You've already bought the extra radiator so knock yourself out and put it betwen the blocks if you want but you might actually make the overall efficiency worse! The additional cooling offered by the extra radiator may reduce the temperature of the coolant entering the MCW-60 but the increased resistance of the extra 120.3 radiator will reduce flowrates and might affect overall cooling efficiency. My prediction is that you will soon be putting a 120.3 radiator on the Members Market when you find that your extra rad is making no appreciable difference to temperatures, has increased the amount of noise (added fans) and increased your loop's complexity.
 
yes i know another triple will make my flow rate suck, but the XSPC rad has very low resistance for a 120.3.

I bought it knowing that, and it will be mounted horizontally to minimise resistance.

edit: as for the noise, i have a T-balancer now to turn all fans of when surfing/movies, complete passive at idle 45c.
 
Last edited:
What difference to resistance will mounting it horizontally make? For every up in a loop there is a down. You appear to have some dangerous misconceptions.

Your system is stable
Your temperatures are very good

Do you just have money to burn? If so, send me some. I can't for the life of me understand why you want to fit this radiator. If the real answer is "because I can" then there's no point asking for opinions on here.

Sorry if I seem rude but I really don't get what you're doing or why, having asked for advice, you're not taking it.
confuse.gif
 
MikeTimbers said:
You've already bought the extra radiator so knock yourself out and put it betwen the blocks if you want but you might actually make the overall efficiency worse! The additional cooling offered by the extra radiator may reduce the temperature of the coolant entering the MCW-60 but the increased resistance of the extra 120.3 radiator will reduce flowrates and might affect overall cooling efficiency. My prediction is that you will soon be putting a 120.3 radiator on the Members Market when you find that your extra rad is making no appreciable difference to temperatures, has increased the amount of noise (added fans) and increased your loop's complexity.
Click to expand...

Agreed - why on earth you would want two triple radiators on one loop is anyone's guess.
 
Jokester said:
Flow rate doesn't make that much difference either to be honest, as Cathar's tubing size testing showed. For a typical heat dump of 200-300W for a multi block loop the delta T across a radiator is about 0.2-0.3C.

It's not worth getting hung up on where to put it in the loop. Put it where it minimises bends and tube length.
Click to expand...

I thought there wasn't a huge difference in flow rates between the tubing sizes - I thought it was just that the difference in resistance from different tubing diameter was pretty small compared to the resistance of water blocks...
 
BornEvil said:
yes i know another triple will make my flow rate suck, but the XSPC rad has very low resistance for a 120.3.

I bought it knowing that, and it will be mounted horizontally to minimise resistance.

edit: as for the noise, i have a T-balancer now to turn all fans of when surfing/movies, complete passive at idle 45c.
Click to expand...

As said, mounting horizontally makes no difference.

You'd probably do better to make a separate loop for the GFX and the XSPC radiator. Adding a BIX2 120.2 to my 10wDDC/CPU/GFX/pa120.2 loop didn't make much difference. How big's the case it's going in?
 
What i mean is that with horizontally mounting, the thread will be left/right instead of up/down.

I am adding another radiator because my CPU overclock is no longer stable after i added the GPU block. temps raised to 78c max and was 71c max before.

It is going in a P182, triple cutout on top, and a swiftec radbox in the back (so i can always take it off if does not work). The radiator only cost £15 (used).

I am by no means new to watercooling and it is a matter of trial and error. I am very happy to accept opinions but it does not take long to physically try it my self. I might well need another loop but it cost nothing to try. it is bleeding atm, will get result soon.
 
Monkey Puzzle said:
I thought there wasn't a huge difference in flow rates between the tubing sizes - I thought it was just that the difference in resistance from different tubing diameter was pretty small compared to the resistance of water blocks...
Click to expand...
Not so much, going from 1/4" ID tubing to 1/2" ID resulted in a 40%+ increase of flow rate, but only a 0.5C decrease in load temps. As long as you were above a certain value (Cathar I think says about 2lpm) any additional flow has little affect on temperatures, the 1/4" ID tubing (6mm) achieved a flow of 4lpm so it was well above that limit.
 
You must log in or register to reply here.
Back
Top Bottom