Sign in with
Sign up | Sign in
Your question

Water Cooling Setup

Last response: in Overclocking
Share
March 3, 2012 2:57:21 PM

Hi everyone, what do you think of this setup?



I had the res next to the pump first (see image below), but it was recommended to place it directly above the pump, so I changed it to the setup in the above image.



Would love to get some advice from you.

Thanks!

More about : water cooling setup

a b K Overclocking
March 3, 2012 3:27:14 PM

^good setup.if it's 800D then you can fit a 360 rad at top.
m
0
l
March 3, 2012 3:40:27 PM

It's a 650D and unfortunately, only 240 rad will fit in there. And it has to be a slim one, otherwise it will hit the ram.

Thanks!
m
0
l
Related resources
March 3, 2012 4:47:17 PM

Hmm to be honest, I don't like his setup. He has water going from the GPUs to the CPU without cooling it. I'd rather run it through another rad before going to the CPU.

EDIT: where is the pump in his case?
m
0
l
a c 190 K Overclocking
March 3, 2012 5:06:54 PM

Haven't clicked the link as I'm on the phone, but you could always alter the tubing to run the rad as you wished,
W/c often includes modding as par for the course :-)
Moto
m
0
l
March 3, 2012 5:10:24 PM

I think he has one of those res/pump combos.

And nice sig, fits right in with what you said.

EDIT: actually it seems he has a XSPC DDC 5.25 Dual Bay Reservoir. Where is his pump?
m
0
l
a c 190 K Overclocking
March 3, 2012 5:17:54 PM

It will be inside the res then :-)
Moto
m
0
l
a c 190 K Overclocking
March 3, 2012 5:29:30 PM

Kind of, the picture looking through the viewing slot from the front, I can't see a pump,
But the pic above, you can see a black thing inside the res, look through the slit in the drivecage,
Plus there are two tubes going to the res and nowhere else for the pump to be, trace all the tubes and they go 'somewhere', pump must be in there
Moto
m
0
l
a c 330 K Overclocking
March 3, 2012 7:47:25 PM

Quote:
Hmm to be honest, I don't like his setup. He has water going from the GPUs to the CPU without cooling it. I'd rather run it through another rad before going to the CPU


Loop order does not matter at all when it comes to system temps. Component temp reading and water temp are not the same...there isn't any 'hot water' in a water cooling loop...if the delta is right, it shouldn't even really get to lukewarm.
m
0
l
March 4, 2012 1:07:57 AM

Sorry, what do you mean by "if the delta is right, it shouldn't even really get to lukewarm". I am not sure I get this.
m
0
l
a c 330 K Overclocking
March 4, 2012 1:12:39 AM

The delta-T calculation for your loop...I go over this in the watercooling sticky with some information on how to determine an estimate for your loop based on several criteria.

Basically, delta-T is a derivative of ambient room temperature (which you can never cool below with normal air or watercooling), heat output by components expressed as watts, water flow rate, radiator cooling ability in watts based on fan speed and total radiator surface area. A delta of 10C would mean your ambient room temperature is 25C and your water temp at 100% load reaches no more than 35C. This number is also different than your hardware temp readings which is taken at the IHS of your CPU/GPU by on-chip sensors and reported. While they aren't the same, there is a correlation between the two.
m
0
l
March 4, 2012 1:14:09 AM

rubix_1011 said:
The delta-T calculation for your loop...I go over this in the watercooling sticky with some information on how to determine an estimate for your loop based on several criteria.


Alright, will check that out. Thanks!
m
0
l
March 4, 2012 5:58:06 AM

rubix_1011 said:
The delta-T calculation for your loop...I go over this in the watercooling sticky with some information on how to determine an estimate for your loop based on several criteria.

Basically, delta-T is a derivative of ambient room temperature (which you can never cool below with normal air or watercooling), heat output by components expressed as watts, water flow rate, radiator cooling ability in watts based on fan speed and total radiator surface area. A delta of 10C would mean your ambient room temperature is 25C and your water temp at 100% load reaches no more than 35C. This number is also different than your hardware temp readings which is taken at the IHS of your CPU/GPU by on-chip sensors and reported. While they aren't the same, there is a correlation between the two.


Totally true. Nice work rubix!

And your other comment about loop order I so true. Wish more people would understand this as the water reaches a balanced temperature throughout the loop.
m
0
l
a c 190 K Overclocking
March 4, 2012 8:34:20 AM

Got to love physics hehe :) 
A lot of people like to put certain components in certain orders though, I think it gives a sense of 'doing it right'
my loop is designed for minimal visibility, but the extra tubing involved just means it takes a little longer to reach thermal equilibrium, it doesn't affect temps just because you have more water in there, or the order it moves around in
the important thing to me is that it moves at a steady rate and is cooled as much as possible on it way :) 
**Edit, and that you can't see much of the tubing etc of course :p )
Moto
m
0
l
March 4, 2012 8:49:19 AM

Well i put into practice on my loop to pull from rad to pump then to cpu to avoid the pressure drop when the water hits the first block which affects flow rate and ultimatly the final temp of the cpu the higher the flow rate at the cpu the cooler it will be.Also the pump stays cooler pulling straight from a rad..
m
0
l
a c 190 K Overclocking
March 4, 2012 9:44:30 AM

However, some would swear blind that a pump should 'always' be next in line from the res,
having the pump in a res negates the pump cooling issue (at least for me)
I am finding more and more that what works in one setup may not be the optimum set up for an identical build
there are always variables, and the kind of folks we are by nature makes us want to tweak and get the absolute best we can from a loop,
we are cursed hehe :) 
Moto
m
0
l
a c 330 K Overclocking
March 4, 2012 2:03:28 PM

Pump after res usually allows for easier filling and bleeding during loop filling. As CPU blocks are the most restrictive in a loop, it is very common to see the pump outlet going straight to CPU inlet. This does help address the pressure drop issue that robustus64 mentions and allows the CPU block to perform at it's best. Most current-gen CPU blocks rely on the high pressure to perform the way they do, instead of older blocks simply loving high flow rates and less restriction.
m
0
l
March 4, 2012 2:20:06 PM

@rubix_1011

I read the whole thing and it has certainly given me a better idea of what water cooling is about. At the same time, there are some parts that I don't completely understand and there are some parts that scare me a little bit.

The parts that I don't understand completely shouldn't be a problem since there is a lot of information in that sticky and I can't process it all at once. I need to research a little more and read it a couple times before I get into water cooling. It is challenging, especially since I have never done it before, but that makes it only more fun.

The part that scares me a little is the water or coolant that's needed. You have presented a great deal of information, though I can't rely on one source only and have to google certain things myself. This is the area where opinions seem to be mixed.

There are people that run into trouble using distilled water + additives, yet again, they are not sure what's the cause of it. Others on the hand seem to be doing just fine with distilled water and killcoil or PT nuke.

For instance, reading through this thread (and some stuff on google) gives me mixed opinions. This uncertainty is what makes watercooling a bit scary and the thought of damaging my hardware is, well, not good.

http://www.overclock.net/t/1205823/silver-kill-coil-not...

I hope you can understand that this concern I have is part of my research and I am trying to get as much information as possible before I get started with it.
m
0
l
a c 330 K Overclocking
March 4, 2012 5:23:03 PM

Corrosion like that has nothing to do with the silver coil but more to do with how well (or how poorly) a component is plated, and with what. If you don't need all that shiny nickel plated stuff, then you are less likely to encounter issues with plating. If your components aren't nickel plated and are still corroding, then that is likely an issue of the quality of brass being used (fittings for example).

I understand the concern, and in reality, every loop is going to corrode at some point in time, but it really depends on the quality of components in the loop. If you are that concerned with corrosion, use a glycol mixture in your loop like Swiftech's Hydrx or even some automotive antifreeze mixed at about 30%.
m
0
l
!