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

Watercooling Build. Dual loop thoughts.

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

SeeThruHead

Member
Joined
Feb 23, 2011
Location
Toronto, Canada
4x 120.3 Ascension Build

Turning this into more of a build planner/build log now that I've settled on parts.

Part // Price // Quantity // Total
Laing DDC Rev. 3.25 Liquid Cooling Pump - (12 Volt -18 Watt) // $64.99 // 2 // $129.98
EK-DDC Dual TOP V.2 G1/4 - Black Acetal // $62.07 // 1 // $62.07
EK-Multioption LINK 1×G1/4 // $11.05 // 1 // $11.05
EK-Multioption Tube 140mm // $19.88 // 1 // $19.88
EK-Multioption TOP W/Thread // $9.11 // 1 // $9.11
Danger Den Fillport - Black // $13.99 // 1 // $13.99
Swiftech MCR-320 Quiet Power Series Radiator - Matte Black // $49.95 // 4 // $199.80
Gentle Typhoon 120mm Silent Case Fan 1850RPM // $15.99 // 18 // $287.82
Tygon A-60-G Norprene 1/2" ID (3/4" OD) // $3.00 // 15 // $45.00
Bitspower Black Sparkle G1/4" Barb Fitting - 1/2" ID // $1.00 // 8 // $8.00
Bitspower 1/2" 90 Degree Rotary Compression Fitting G1/4 With BARB // $11.99 // 4 // $47.96
Bitspower 1/2" 45 Degree Rotary Compression Fitting G1/4 // $10.99 // 5 // $54.95
Lamptron Elite Aluminum Reusable 3/4" OD Liquid Tubing Clamps - Anodized Silver // $3.99 // 13 // $51.87
Koolance QDC (Extreme Flow) No-Spill, Female Compression (13mm, 1/2" x 19mm, 3/4")(VL4N-F13-19S) // $15.99 // 1 // $15.99
Koolance QDC (Extreme Flow) No-Spill, Male Compression (13mm, 1/2" x 19mm, 3/4")(VL4N-M13-19S) // $14.99 // 1 // $14.99
Indigo Xtreme™ Engineered Thermal Interface (ETI) Kit for Core i5, LGA 1156 - 2 Pack // $19.99 // 1 // $19.99
Arctic Cooling MX-4 Super High Performance Non-Conductive Thermal Compound (4g) // $11.99 // 1 // $11.99
CPU water Block // $75.00 // 1 // $75.00
GPU water Block Fullcover // $105.00 // 1 // $105.00

Ascension CYO (Choose Your Own)[ASCENSIONCYO] // $529.00 // 1 // $529.00
Lamptron FC Touch - 30W - 6 Channel Aluminum Rheobus w/ Touch Screen - Black (FCT) // $74.99 // 1 // $74.99
Lian Li 5.25" Multi-Media I/O Ports - Black - (BZ-U02B) // $28.99 // 1 // $28.99
Lian Li 5.25" to 4 x 2.5" HDD Converter - Black (BZ-525B) // $24.99 // 1 // $24.99
Lian Li EX-H22B SATA Hot Swap HD Cage - Dual 3.5" Drives - Black // $64.95 // 1 // $64.95
MDPC SLEEVE-KIT // $50.00 // 2 // $100.00
Accessories - Nexus Anti-Vibration PC Mounting Kit - 60 components // $14.95 // 1 // $14.95

Other:
Components:
Vertex 3 SATAIII SSD x2
WD Caviar Black SATA II 2TB x2
Whatever X68 Sandybridge-E Mobo is best
Whatever X68 Sandybridge-E Proc is best
8 gigs 2400mhz ram
24l9b21.jpg rvzj8k.jpg
 
Last edited:
You don't need push/pull fans. Your raddage size won't need that much airflow. Unless you plan to be at 800 RPM on the fans, very quiet. 1000RPM on those fans is still very quiet.

You will have lots of problems putting 6 GT1850s on one fan header. Might be too many amps.

WHICH MCP655? Vario/non vario? Don't matter much, you would be fine with either.

You want auto fan control? Pick your fan controller carefully. Actually, you have 3 GT1850's at 1000 RPM and leave them on the CPU loop, they don't need adjusted at all, temps will be great at load. Same with the 580 on the 120x2 radiator. Quiet.
 
So you're saying I should skip the 3 top mounted fans and just run 3. The Gt's seem to be pretty low amperage. Website says 0.083A How would I run them at 1000 without a fan controller? (I think 1450 would actually be the sweetspot for me but I'd like the ability to ramp it up to 1850 incase I need it, automatically adjusted of course)

EDIT: Pump would be the vario version. I like having options. The option to turn it down a notch if it's too loud seems good to me.
 
Last edited:
3 of those fans work great on one channel of many quality controllers. Think the one you selected is fine. I have the Lamtron FC6. Two channels have 3 fans for each rad. Parellel, one rad at 970, another at 1020.

0.083 is running speed, you really missed the boat by not looking at starting current. I have posted on it before. You can find it. Scyth is made by Nidec. You can do the searching.

You take the time and read a few umm manyx10 posts here? About the size of the D5? LOL, fitting it all, with TWO of those pumps in that case....

You ever hear of build logs? We have had many nice builds here with that case. Ever hear of a resiviour that can hold two pumps? Or a pump by itself would be enough for your loop? No need for two loops if your a gamer if you really understand your heatload?

You got a great start, your not jumpin into it. Awesome. But you can make some 'inform yourself' changes that will help. Takes nothing but time, a week or so.

Please read lots of posts here and ALL the way through. The 'light bulb' in your head might not come on till reply 23. And you could use 3-9 'light bulbs'.

Good start! Get to work. A link that might help, you might be past most of this already.
http://www.overclockers.com/forums/showpost.php?p=6489396&postcount=3
 
You could always go for a dual ddc res, I know XSPC makes one and koolance has a new one that just came out.
 
You could always go for a dual ddc res, I know XSPC makes one and koolance has a new one that just came out.
That actually sounds like a great idea.
I wouldnt mind making something similar to this biuld here:
 
Last edited:
The name of the game here for me is future proofing. I know complete future proofing is impossible but I want to be prepared in case I need more cooling capacity at some point. The way I have things set up now adding SLI to the 240 loop isn't going to provide me with the best loop efficiency. Main thing holding me back is space. So I've been looking into Mountain Mods cases and the really nice build by shazza that was in Conundrum's link. Although I'm a little miffed why the only panel with dual 360 support is the front panel, I might just go that way anyway. Updated build plan later.
 
I think I'm on the right track here but I have some questions that have popped up from what I've read over the past week. Seems like I want to go single loop. Reason for that being that I probably won't be loading both the cpu and gpu at the same time, and I want my radiators potential to be maximized when loading either, and not sit almost unused on the loop with less heat. I've heard that lots of blocks are flow limited. I assume this means they have a set flow rate or maximum flow that they will allow through, it would make sense to me to then put all the blocks in parallel. This would essentially add the flow limits of the two blocks together would it not? I'm unfamiliar with how water flow works in a system like this though. Would it be possible that one block has a high enough flow rate compared to the other that the water would choose the path of least resistance and only flow through one block? Or negatively impact flow to the other block to a noticeable degree?
 
Last edited:
A single pump should be fine unless you need redundancy or are using very restrictive blocks. The idea is to keep flow at 1gpm or higher. Anything over 1gpm has very little effect on temps. Running the entire loop in serial will probably have the best results and make it easier to run tubing. The other option is to run a bay res with pump that can accommodate two loops and then put your rads in one loop and the blocks in the other loop. The water in the res will get mixed fast enough to make sure your cooler water from the rads is going to the loop with the blocks. Something like the DD Monsoon or the Koolance bay res would accomplish this.
 
The other option is to run a bay res with pump that can accommodate two loops and then put your rads in one loop and the blocks in the other loop. The water in the res will get mixed fast enough to make sure your cooler water from the rads is going to the loop with the blocks. Something like the DD Monsoon or the Koolance bay res would accomplish this.

That sounds outright foolish to me honestly. There would be no advantage to this as you would have one incredibly low restriction loop and one incredibly high restriction loop (relatively speaking)
 
I think i'm going to shy away from running things in parallel. Going to be too messy tubing wise. I want to over rad. That's my main thing here. The possibilities open up in my head when I over rad. I can add more heat load at a later time, I can run quieter fans and still have enough heat removed from the loop etc etc. My case is going to have room for 4x360 rads. I would use all of them at some point (probably not right away but it presents a nicely modular upgrade path) Now I'm looking for what is going to allow me to keep a high flow rate in the loop, (2.0 gpm is what I'm going for) with 4 360 rads, (good ones, and these all seem to be fairly restrictive) and 1-3 blocks. (1cpu 1gpu to start and possible another gpu at some point. I'm thinking I'll need 2 pumps in order to get enough flow for proper cpu block to funtion. does it matter where they are in the loop? my common sense tells me they should, one before the blocks and another before the radiators, but I may not know enough about the physics here to be right in that assumption.
 
You want the pumps back to back so that the backpressure on the one is alleviated by the other (and vice versa)
 
I'll admit I'm having a bit of trouble understanding. Is there a good thread you can recommend I read? So you're saying I should run them directly after one another? How does this accomplish great flow rate?
I would think in a setup like this resevoir > pump1 > 2x water blocks >pump2 >4x radiator

that the pressure would drop as it goes through the water blocks and then the next pump would increase pressure again pushing the water through the radiators. You setup would be a lot easier and cleaner to implement but I'm miffed as to the physics here.

Anyway, putting them back to back simplifies my whole setup and tubing problems. Thinking of using 2 mcp355's with this setup (ek dual top and res)
ex-pmp-95_2.jpg
It gives me the option of running the input to the resevoir and a fillport line out of the top of the res which is exactly what I wanted.
EDIT: Updated my main build information, still undecided which rad's to go for. One is very cost effective but the other gets me much better performance if I should ever need it. But for almost triple the price. (the swiftech can be found for 50 dollars....)
Edit2: Can anyone recommend a good 360 shroud? I'm having trouble finding anything that isn't really expensive because it's ugly *** acrylic. 10 dollars a shroud would be nice. I might end up just using hollowed out fan housings but i think a fully open 360 shroud for each rad (or even a 1080 shroud :D) would be better and cheaper (i dont have any spare fans lying around)
 
Last edited:
That sounds outright foolish to me honestly. There would be no advantage to this as you would have one incredibly low restriction loop and one incredibly high restriction loop (relatively speaking)

Yes, one loop would be more restrictive than the other, however it would make running the tubing easier and as long as both loops are at 1gpm or above it really wouldn't matter. As far as being foolish, it's more foolish to dismiss something you have never tried. I might be new here, but I have been doing this for awhile and have built a few systems using water cooling and to dismiss something I say without knowing me is rather foolish, just me being honest and all.
 
i think running is series is going to be my best bet. I dunno the physics behind determining whether or not your suggestion actually makes sense. But regardless the tubing for it is more complicated than series and what i want to get into.
Right now I'm looking for a way to keep 2gpm flow through 4 rads a few 90 degree fittings and 3 blocks. with the cleanest tube routing possible.
you might say four rads?!?!?! that's overkill. But i want the option for silent fan operation, and also the ability to turn up the fans and cool 3 overclocked gpus and an overclocked sandy bridge E when it comes out.
 
Last edited:
I'll admit I'm having a bit of trouble understanding. Is there a good thread you can recommend I read? So you're saying I should run them directly after one another? How does this accomplish great flow rate?
I would think in a setup like this resevoir > pump1 > 2x water blocks >pump2 >4x radiator

that the pressure would drop as it goes through the water blocks and then the next pump would increase pressure again pushing the water through the radiators. You setup would be a lot easier and cleaner to implement but I'm miffed as to the physics here.

Anyway, putting them back to back simplifies my whole setup and tubing problems. Thinking of using 2 mcp355's with this setup (ek dual top and res)
ex-pmp-95_2.jpg
It gives me the option of running the input to the resevoir and a fillport line out of the top of the res which is exactly what I wanted.
EDIT: Updated my main build information, still undecided which rad's to go for. One is very cost effective but the other gets me much better performance if I should ever need it. But for almost triple the price. (the swiftech can be found for 50 dollars....)
Edit2: Can anyone recommend a good 360 shroud? I'm having trouble finding anything that isn't really expensive because it's ugly *** acrylic. 10 dollars a shroud would be nice. I might end up just using hollowed out fan housings but i think a fully open 360 shroud for each rad (or even a 1080 shroud :D) would be better and cheaper (i dont have any spare fans lying around)

I have used that pump top in the past and it will increase your flow, but mainly it's for redundancy in case a pump fails. As far as rads go, the Swiftech is the staple of the water cooling community for a reason, it's pretty cheap and the rads, like feser that are twice as much, offer very little gain in comparison. However, if you would like something in between price wise with performance comparable to Feser, than look at the XSPC RX series rads or even EK and Koolance. Shroud wise you only have a few options.

http://www.performance-pcs.com/cata...e=product_info&cPath=59_335&products_id=28827

http://www.performance-pcs.com/cata...e=product_info&cPath=59_335&products_id=28832

http://www.performance-pcs.com/cata...e=product_info&cPath=59_335&products_id=25684

http://www.performance-pcs.com/cata...e=product_info&cPath=59_335&products_id=24275

Or you could go with Thermochill and a shroud from them:

http://www.aquatuning.us/product_in...l-TA120-3-Triple-Radiator---360-Radiator.html

http://www.aquatuning.us/product_in...l-TA120-3-Triple-Radiator---360-Radiator.html

And finally, just buying some cheap fans and gutting them. Usually tou can get them for around $3 each.
 
i think running is series is going to be my best bet. I dunno the physics behind determining whether or not your suggestion actually makes sense. But regardless the tubing for it is more complicated than series and what i want to get into.
Right now I'm looking for a way to keep 2gpm flow through 4 rads a few 90 degree fittings and 3 blocks. with the cleanest tube routing possible.

Why 2gpm? Just curious as to where you got that number and why. Tests have been done and anything over 1gpm only have minuscule benefits. Both Skinnee and Martin have done tests on this along with various people over the years. All I am saying is don't get caught up in flow.

As far as the dual loop with running the rads on one loop and the blocks on the other, if it complicates your tubing runs then don't use it. Go with what makes sense and simplifies the setup. I have the same MM case and my rads are all on the front, so I come out of DD Monsoon to the first rad and then the second and then the third and back to the res, it was a simple loop. My second loop goes to the cpu, then the GPU's and back to the res. It was the easiest and simplest way for my setup, but if it doesn't work for you then I wouldn't do it.
 
i added a little diagram i drew to my original post, detailing how i want to set everything up, tube routing is really very easy and going to be quite clean looking in that setup. I've never seen it done your way so I'm just going to shy away from that. I haven't done the testing nor i have the technical knowledge to discount it however. Those shrouds are all garbage it seems. not what I'm looking for. i'm gonna look around town and see if i can have something machined up. who knows, will probably end up going with gutted 1 dollar fans from semewhere.
As for flow rate i was looking at this http://www.overclockers.com/forums/showpost.php?p=6332602&postcount=4
although misremembered the actually recommended flow rate.

This is bothering me "You reduce flow and you can severly impact the cooling capability of the CPU block" With so many rads and blocks the restriction would add up. I want to make sure i'm hitting at least 1.5 gpm to get the proper cooling out of any cpu block that i want to use.
 
STH, your getting the idea. Your flow rate needs to be high enough, yes. Two modern pumps in series is plenty for your loop. Two DDC 3.2 or two D5 varios would be fine, in series, right next to each other.

I don't understand the need for 4 rads, but okay. Rads are a very very small consideration for flow calculations. They are very unrestrictive. You'll have a TON of fittings etc, you have no clue yet how hard it will be to build the setup you envision. I bet over $200 easy in just fittings.

One thing to consider, how will you setup the 4 rads? All pulling cool room air into the case? How will you exhaust it? Having a case with 4 of those rads 'MIGHT' impact your mobo, hard drive etc cooling. You still need good airflow.

For example, two rads pulling warm air into the case and shausting the air through two rads gives you two complexities.

1. Airflow in the case will be reduced because rads slow airflow, a LOT.
2. The two rads getting warm case air will be less efficient as they pul the HOT air from the case. There still are heat producing parts in the case that need cooling.
 
There are 9 intake holes on the front of the case. I actually have drawn a diagram up at the top there. Thats three rads pulling in cool air from outside the case. One rad exhausting. In addition to that there is 4 unrestricted 120mm holes for exhaust. 2 on the back and two on the top. Or alternately i could get a third hole added to the top for 5 unrestricted exhaust.
From this post here it looks like the gentle typhoons have about 25 CFM when running through that rad at 1450rpm (i dont know where there might be info for my specific radiator choice but i'm just going to go with that number for now)

The specifications for the gentle typhoon 1450 rpm version say 50cfm. I'd assume the cfm would be close for a 1850rpm running at 1450 rpm via fan controller.
So thats 9 intakes at 25cfm 225 cfm of intake
3 exhaust at 25 cfm and 4 at 50 cfm
275 cfm exhaust.
I think airflow would be fine (would it be better to try and match intake/exhaust cfm more closely?)

now i understand that ALL my intake is being heated by the radiators. the one exhaust rad will be less efficient but i think it's still going to add enough heat dissipation to warrant the addition.
Do you guys think that there will be enough airflow, even at its warmer temps to cool the mother board? Keep in mind this is all being designed for Sandy Bridge E processors which i believe will have the NB function built into the chip, and therefor cooled with the cpu.

Also I will be adding heatsinks to everything on the board that does not have a heatsink and may need it.
EDIT: there will also be somethign like this cooling the ram:
corsair-dhx-dominator-ram-cooler-4.jpg
Which should add some airflow to the ram and even modules that are situated around the ram.
Keep in mind I will be buying one of the next enthusiast series motherboards which should have fairly good motherboard heatsinks already on it.
I understand fittings are going to be expensive. There will be 6 90 degree adapters (72 dollars), 2 45 degree adapters (22 dollars). 15 barb fittings (15 dollars), 15 lamptron clamps (60 dollars),
and one fillport (14 dollars) for a total of 170 bucks before tax. (fillport was not included on the diagram but will connect to the top of the reservoir)
I'm also considering adding a t line splitter with the second end terminating in a male quick disconnect. I would then have the female end of the quickdisconnect attached to a spare length of tubing which would enable me to easily drain the system. (air intake through the fillport and water going out the quickdisconnect) without having to actually take anything fittings off.

EDIT: about the fan setup, will the air take the path of least resistance? What i'm worried about having more exhaust cfm than intake is that the 225 cfm of intake will split up in the exhaust fans like so: 200cfm out of the unrestricted fans, and 25 cfm out of the 3 fans on the exhaust radiator. If that is the case then it would be beneficial to reduce the 4 unrestricted fans to a combined 150CFM of exhaust so there is enough intake CFM to feed air to the 75cfm of the exhausting radiator. Thoughts?
 
Last edited:
Back