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

Newbie water-cooling system

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

MadPug

Registered
Joined
Mar 16, 2016
Hi all, I'm new here and this is the first ever computer forum I've joined. I'm looking to build a brand new (also first water-cooled) pc to run the Asus PG279Q 1440p 165hz G-sync monitor.

Specs:
Asus Maximus VIII Formula motherboard
Intel i7-6700K
Nvidia GTX 980Ti
Corsair Dominator Platinum CMD16GX4M4C3200C15 (4x4GB)
Cooler Master V850 80 Plus Gold PSU
Samsung 850 Pro 256GB SSD

Water-cooling components that I've decided on so far:
EK Supremacy Evo Elite Edition 1151 CPU block
XSPC Razor GTX 980 GPU block
EK Full cover GTX 980 Nickel backplate

The reservoir and pump will depend on the case, I was going to get the Phanteks Enthoo Evolve ATX case because the tempered glass version looks extremely sexy, however it is too small and the exhaust doesn't look very promising. So the most likely candidate right now is the Enthoo Primo.

With this huge case, I'm thinking of running 2 separate loops for the CPU and GPU, using the top to mount the GPU radiator and the right side panel to mount the CPU radiator, they will both be pushing air out of the case. For intake fans there will be 2 in the front, and 2 in the bottom, there will also be a single exhaust fan on the rear.

What do you guys think of this set-up? Any flaws? I avoided mounting a radiator in the front because I don't want hot air inside the case, so both radiators will be mounted at exit points, is this a good approach? The bottom intake fans also provide direct cool airflow to the top mounted radiator to push hot air out of the case. The top radiator will most likely be a 360 and the side radiator will be a 240.

As for a pump, is D5 always better, I watched one of JayzTwoCents' video and he said although D5 has a lot of air flow, it doesn't have as much head pressure. So if I make my loop as simple as possible, and with one loop only going through one block, is the D5 a better choice?

Thanks guys.
 

Attachments

  • Primo-6.jpg
    Primo-6.jpg
    69.3 KB · Views: 340
  • Primo-4.jpg
    Primo-4.jpg
    58.3 KB · Views: 343

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Welcome!

As far as 2 loops... its not remotely needed with 2 parts in it and a D5. With a 980 and CPU in the loop, you could easily run this on a single 3x120mm radiator with overclocking them both. ;)
 

Blaylock

"That Backfired" Senior Member
Joined
Jun 5, 2013
Location
Go Blue!
Welcome MadPug! I agree plus generally speaking dual loops can have more issues. Twice as many parts to fail.
 

ITAngel

Member
Joined
Aug 1, 2012
Location
Wyoming
Hi there welcome! That is a nice case I wish I could have gone that route before but instead I been shrinking mine. lol, anyways. I was thinking the same thing on what EarthDog mention. I think the single rad & loop would be cleaner and much more simple that way than like mine trying to run dual 240mm rads. ;) I agree dual loops might not be the best but it looks cool. I like the idea of the Primo case with dual rads one top and one on the bottom.
 

Witchdoctor

Member
Joined
Apr 12, 2014
I am not even close to as conservative than my two friends here.
While it is certainly true you could easily run this off a 3.120 it is also just as true as you can have as many radiators as you so choose.
Depending on the delta you are looking for and how much sound you are willing to put up with are things you have to think about.
Personally I use 70 watts per 120 as a near silent solution , 90 watts as tolerable and 110 watts to loud for my liking

So in my way of thinking based upon pure experience, not any mathematical conclusions

1 loop

341 w / 70 w = 4.87 that tells us 5.120 is near silent with a 8c ish delta

341 w / 90 w = 3.78 that tells us 4.120 is noticeable with a 12c ish delta

341 w / 110 w = 3.1 that tells us 3.120 is loud with a 16c ish delta






Obviously this this is all BS and I have no idea what I am talking about, but it looks reasonable. :rofl:

I would consult the water cooling stickies as they are experts , jz is entertainment

http://www.overclockers.com/forums/...Water-Cooling-Your-PC-***-READ-THIS-FIRST-***


The math don't lie and what you can learn along the way will help you have a fun and rewarding water cooling experience

Best of luck and welcome to OCF :thup:
 
OP
M

MadPug

Registered
Joined
Mar 16, 2016
Thanks for the input everyone, after second though I do agree as a beginner I should stick to one loop for now, maybe 2 loops when I felt the need to go SLI in the future. Also thanks for reminding me about the article Witchdoctor, I had a brief look at the article before but hadn't read the radiator part. But can I just clarify, "so if running an i7, NB block and single GPU, use at least a 120.4.", what is a NB block?

I'm actually reconsidering the Phanteks Enthoo Evolve ATX case because of its unmatched build quality, i.e. thick aluminium panels all around. It's a lot smaller than the Primo and the performance is a lot worse in terms of water cooling but I can't take my mind off it, so I'm trying to make it work.

Evolv-ATX-6.jpg

240 for the i7 and 120 for the 980Ti, I tried to not add a radiator on the front because 1) to allow more space for reservoir/pump and not make the whole system look so crowded, and 2) to avoid having not air in the case. Also, because the case is only 495mm tall, I'm thinking about using EK's slim rad (26m) with a FPI of 22 and sacrificing silence for the aesthetics of the case. Will this single 360 slim rad be sufficient to provide decent cooling in your opinion?

Looking at this set up below, doesn't the front rad reduce cooling performance because it's blowing the hot air in the radiator into the system and feeding it to the top rad?

Phanteks_Enthoo_EVOLV_ATX_35.JPG
 

ITAngel

Member
Joined
Aug 1, 2012
Location
Wyoming
Thanks for the input everyone, after second though I do agree as a beginner I should stick to one loop for now, maybe 2 loops when I felt the need to go SLI in the future. Also thanks for reminding me about the article Witchdoctor, I had a brief look at the article before but hadn't read the radiator part. But can I just clarify, "so if running an i7, NB block and single GPU, use at least a 120.4.", what is a NB block?

I'm actually reconsidering the Phanteks Enthoo Evolve ATX case because of its unmatched build quality, i.e. thick aluminium panels all around. It's a lot smaller than the Primo and the performance is a lot worse in terms of water cooling but I can't take my mind off it, so I'm trying to make it work.

View attachment 176897

240 for the i7 and 120 for the 980Ti, I tried to not add a radiator on the front because 1) to allow more space for reservoir/pump and not make the whole system look so crowded, and 2) to avoid having not air in the case. Also, because the case is only 495mm tall, I'm thinking about using EK's slim rad (26m) with a FPI of 22 and sacrificing silence for the aesthetics of the case. Will this single 360 slim rad be sufficient to provide decent cooling in your opinion?

Looking at this set up below, doesn't the front rad reduce cooling performance because it's blowing the hot air in the radiator into the system and feeding it to the top rad?

View attachment 176898

That is an amazing case too, I am getting today the Enthoo Pro M to try do simple water cooling but someday I can make something like this here.

source http://www.overclock.net/t/1587849/...tercooled-and-modified-enthoo-pro-m-completed
900x900px-LL-11bda5a3_2016NevermoreRavenwolfPCPPKurrgenOCN2.jpeg
 

Blaylock

"That Backfired" Senior Member
Joined
Jun 5, 2013
Location
Go Blue!
FTR the air passing thru the rad isn't that hot. Only a few degrees above ambient and with proper air flow will not linger inside the case. It really isn't an issue to have a rad up front in taking air.
 
OP
M

MadPug

Registered
Joined
Mar 16, 2016
I have no choice but to go with a thin rad if I go with the Enthoo Evolve ATX (for aesthetics reason), how much static pressure would you say is needed for a 22 FPI rad?

The Noctua NF-F12 runs at 3000rpm and has a static pressure of 7.63 mm/H2O, but also a noise level of 43.5 dBA, do you think it's suitable?

When I look up on fans there are only discussions of AF v SP but not the level of static pressure in relation to FPI, so it's hard to determine what is good static pressure and what is not enough. The Corsair SP120 looks like a popular choice but is 3.1 mm/H2O considered a good amount of static pressure?
 
OP
M

MadPug

Registered
Joined
Mar 16, 2016
For this set-up, the front rad will be a Black Ice Nemesis 280GTS (29.6mm thick, 16 FPI) and the top will be a EK CoolStream SE 240 Slim (26mm thick, 22FPI), what kind of static pressure fans would be suitable for both rads?

Also, assuming all the fans are the same RPM, will this set-up achieve a positive air flow since the front rad is 280 and the top rad is 240? So there will be 3 140mm fans for intake and 3 120mm fans for exhaust.
 

Attachments

  • Evolv-ATX-6.jpg
    Evolv-ATX-6.jpg
    64.1 KB · Views: 247

ITAngel

Member
Joined
Aug 1, 2012
Location
Wyoming
For this set-up, the front rad will be a Black Ice Nemesis 280GTS (29.6mm thick, 16 FPI) and the top will be a EK CoolStream SE 240 Slim (26mm thick, 22FPI), what kind of static pressure fans would be suitable for both rads?

Also, assuming all the fans are the same RPM, will this set-up achieve a positive air flow since the front rad is 280 and the top rad is 240? So there will be 3 140mm fans for intake and 3 120mm fans for exhaust.


That is kind of what I am running right now on air without the rads yet still dealing with my res and pump configuration and placement but I think I came up with a good idea to make it all work out. XD
 

Blaylock

"That Backfired" Senior Member
Joined
Jun 5, 2013
Location
Go Blue!
...what kind of static pressure fans would be suitable for both rads? With higher fin count rads you will need the high static pressure fans. The benefit of lower FPI rads is you can slow the fans down for a quieter set-up. I would still look for fans with a high static pressure as you always have the option to slow them down with PWM or voltage. Some great fans I would consider are Noctua, EK Vardar, Noise Blocker eLoop, etc.

Also, assuming all the fans are the same RPM, will this set-up achieve a positive air flow since the front rad is 280 and the top rad is 240? So there will be 3 140mm fans for intake and 3 120mm fans for exhaust. The fact that you are mixing 120 and 140mm fans means that they will not all be the same RPM. Even if they were the CFM/SP will be different stickly based on size. Add in the fin density difference between the two rads and you end up with to many variables for anyone yo give you an accurate answer. My best guess would bee to add up the CFM of your intake and the CFM of the exhaust and see which is greater. As I mentioned though this is no guarantee as you'll lose CFM trying to push through the rads. To be fully honest with a custom loop there in an included maintenance that must be done regularly. The only benefit to positive case pressure is a slight reduction in dust. Since you'll be doing a tear down every year for flushing/cleaning just blow out the little dust that accumulates. No Worries.
 
OP
M

MadPug

Registered
Joined
Mar 16, 2016
With higher fin count rads you will need the high static pressure fans. The benefit of lower FPI rads is you can slow the fans down for a quieter set-up. I would still look for fans with a high static pressure as you always have the option to slow them down with PWM or voltage. Some great fans I would consider are Noctua, EK Vardar, Noise Blocker eLoop, etc.

The fact that you are mixing 120 and 140mm fans means that they will not all be the same RPM. Even if they were the CFM/SP will be different stickly based on size. Add in the fin density difference between the two rads and you end up with to many variables for anyone yo give you an accurate answer. My best guess would bee to add up the CFM of your intake and the CFM of the exhaust and see which is greater. As I mentioned though this is no guarantee as you'll lose CFM trying to push through the rads. To be fully honest with a custom loop there in an included maintenance that must be done regularly. The only benefit to positive case pressure is a slight reduction in dust. Since you'll be doing a tear down every year for flushing/cleaning just blow out the little dust that accumulates. No Worries.

The highest SP fan I could find was the Noctua NF-12 Industrial PPC 3000RPM, producing 7.63 mm/H2O of static pressure, but is this an overkill? Is something like 3 or 4 mm/H2O plenty enough?

Considering the front rad has a lower fin density (i.e. less restriction) and bigger fans running at the same RPM, wouldn't it be very likely that it'd be positive? I just don't like the idea of having a negative case pressure :-/
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
I just don't like the idea of having a negative case pressure
Why?

First, 'pressure' is really a relative term. Remember a PC case is not sealed so while there may be measureable pressure differences, its truly negligible. you are more airflow on intake than exhaust. I wont argue the cooling merits of more intake CFM than exhaust (I prefer more exhaust CFM than intake), but about the only benefit is keeping dust from coming in the cracks of the case (assuming you have filters on all the intakes). I don't have a particularly dusty house, but, I really don't have a problem with it in the cracks at all.
 

Blaylock

"That Backfired" Senior Member
Joined
Jun 5, 2013
Location
Go Blue!
...and you'll be doing spring cleaning on your loop anyhow.

As far as SP I would think you'd be ok with anything above 2.5mm/H2O for the 16FPI rad and anything in the 3-4 range for the 22FPI rad. This is just a guess as I'm not aware of any chart or formula to determine this.
 
OP
M

MadPug

Registered
Joined
Mar 16, 2016
Why?

First, 'pressure' is really a relative term. Remember a PC case is not sealed so while there may be measureable pressure differences, its truly negligible. you are more airflow on intake than exhaust. I wont argue the cooling merits of more intake CFM than exhaust (I prefer more exhaust CFM than intake), but about the only benefit is keeping dust from coming in the cracks of the case (assuming you have filters on all the intakes). I don't have a particularly dusty house, but, I really don't have a problem with it in the cracks at all.

I just feel that drawing in cool air is slightly more important than pushing out hot air, but then it wouldn't matter in my case because the intake fans would be pushing hot air from the radiator into the case anyway. There's no science behind my preference it's just to make me feel better lol

Also to me, keeping dust from the holes and cracks around the case is very appealing to me.

Why do you prefer more exhaust than intake?

Just another question, will the limited openings on the top of the case significant affect affect the top fans' ability to exhaust air? Will the hot air bounce back into the case because of that?

Evolv-ATX-3.jpg

Thanks very much for helping :)

- - - Updated - - -

...and you'll be doing spring cleaning on your loop anyhow.

As far as SP I would think you'd be ok with anything above 2.5mm/H2O for the 16FPI rad and anything in the 3-4 range for the 22FPI rad. This is just a guess as I'm not aware of any chart or formula to determine this.

Can't go wrong with the Noctua with 7.63 mm/H2O then haha, the 140mm version even goes up to 10.52 mm/H2O. But damn it's a loud fan at 43.5 dBA
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
I prefer exhaust as it removes any 'hot spots' in the case with more air coming out than in. What is the point if getting more air in the case if you can't get it out... and as I said, I do not have dust issues with more exhaust CFM.

Those vents on the side are the only things there for exhaust (outside of the rear fan?!!!), wow, yeah, I wouldn't really call that optimal at all...