• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Water cooling setup

Thrust

Limp Gawd
Joined
Sep 20, 2006
Messages
482
Well I've finally broke down and decided to install my 1st water cooling system (I can thank my nephew who frequents these forums for this. His 2.1 C2D is running faster than my E6600 w/air). Completely from scratch (i.e. heater core, PVC reservior, etc). I've been pondering the water lines & 2 different routing setups.

It seems that virtually every water cooled system I see is much like the example in this pic: http://home.ptd.net/~thrust/loop.jpg

However, it just seems to me that this setup would be even better: http://home.ptd.net/~thrust/2loop.jpg

(The pics are only for reference to show water routing.)

It would seem initially that having nice cool water coming from the radiator going to both GPU & CPU at the same time would be most advantageous (rather than having warmer water exiting CPU block to enter the GPU block). Are there any problems with using a "T" or "Y" and doing this? I'm thinking pressure differences through the GPU block and CPU block might cause an issue but figured I'd talk to you experts here. :p

P.S. Addt'l info: 1/2" lines, Swiftech 8800GTX block/ramsinks, Swiftech Apogee GT and a Via Aqua 1300 (370gph)

Thanks,

']['
 
it seam to be interesting why don t use a secon radiator before the gpu ... like this

Projet1.jpg
 
Why don't you put the radiator after the cpu and gpu? This seems to make the most sense to me.

you have a water resivoir (your cold supply), it gets heated by the cpu and gpu and then gets cooled back down and dumpted back into the resivoir at a cooler temperature.

With the radiator after the resivoir and before the cpu and gpu the system dumps hot water back into the resivoir where it mixes with the cooler water. the temperature of the water only goes up slightly and the radiator doesn't have as much heat to dissapate then the water circulates back to the resivoir heats it up again and so on.

With out some calculations i don't have any proof but it seems less efficient this way.

Thought i would ask, maybe somone here as proven this method more effective, or maybe everyone just follows suit.
 
well, I read a review where the guy moved his radiator from after the CPU/GPU to before it and his temps dropped by 3C. I fully understand your logic but the water would seem to be at it's coldest point as it exits the radiator. Therefore, it seems sensible to have the route of that coldest water go directly into the hardware you're attempting to cool (in this case the CPU). Thanks for the input! Anyone else have some forethought or insight?
 
if you wanna be more effiicent you have do built 2 water loop ... one for the cpu and one for the gpu ...

index.php
 
if you wanna be more effiicent you have do built 2 water loop ... one for the cpu and one for the gpu ...

index.php

That's not necessarily true. One CPU and GPU in a loop should be fine. The temperature differences will be moot for the most point as far as overclocking is concerned.

2 GPUs, 1 CPU, NB, SB, ==Two loops would be good.
 
yeah, I kinda figured it might be moot point to split the tubing but if you're anal, you're anal. LOL thanks for the responses!
 
As far as part positions go, as long as the res is before the inlet of the pump, everythings gravy. For example, my loop is Pump > CPU > Rad > Res> Pump.

3.5ghz (lowered to 500x7) @ 1.5v = 38c idle, 51c load.
3ghz @ 1.3v = 26c idle, 39c load.
1.8ghz stock @ stock voltage = 22c idle, 31c load.
 
Back
Top