Need some configuration advice on a new water loop

[Pschu]

Hi, I am planning on putting together my water loop with in the next week or two. I have a question on how I should configure the loop. I plan on running one 240mm rad and one 120mm rad along with my cpu block and gtx680 block.

My question is:
Should I go
RES->240RAD->120RAD->CPU->GPU->and back to the RES

-or-

RES->240RAD->CPU->120RAD->GPU->and back to the RES
?

What do you think will be better for temps?
Will there be a noticeable gain in CPU temp running one RAD after the other? Or is it worth it to cool the water down after the CPU for the GPU?
Thanks for your time!

 

[rubix_1011]

Have you read through the watercooling sticky?

 

[toolmaker_03]

when i was running this setup i ran it the 2nd way you have up there and the first but i stuck with the second it was not that the system still did not get hot, but it took a hour longer to get there

 

[Pschu]

What I got from the sticky is that it doesn't really matter as to what the order is. To me, I believe it does, because you would be applying more cooling after a main, hot component. I am new to this so this is just a theory. That is why I am asking. Personal experience speaks better than a sticky IMHO. And thanks for your reply toolmaker, I will keep that in mind.

 

[rubix_1011]

There isn't a 'hot' component in a watercooling loop. I wrote the sticky- I've also watercooled for about the last 10 years. I've tried various orders in the loop, it does not matter. I've even linked to some studies in the sticky as to WHY loop order does not matter.

I'm not sure I understand that even though something is written in the sticky, the information 'just might be wrong anyway'?

 

[Pschu]

I apologize rubix, no need to get hostile. But, in tookmaker's post, he did claim it made a difference in the amount of time it took to heat the loop. I am not saying you are wrong, or don't know what you are talking about, I am just curious if there are any differences what-so-ever. Now, may I ask, from your personal experience, did you see any of the same result as toolmaker posted?

 

[Pschu]

I looked at that link closer and I understand that the heat may not be different, but what about the time it takes to heat up?

 

[rubix_1011]

Not being hostile (intentionally) but there are reasons the info exists; I've done it (first because I was uncertain if there really WAS any difference) and others have done it with actual test data to back it up.

Due to the high specific heat and thermal conductivity of water, there will always be much lower reported temps by hardware and your water delta will always be a function of flow rate, fans (airflow rate), radiator surface area and efficiency, ambient room temperature and total amount of heat in watts being introduced into the loop and removed at the same time.

Regardless of where you place water blocks, radiators or pumps, you are still moving water at the same rate, still adding the same amount of heat energy in watts and removing the same heat energy in watts. Differences of time to reach loop delta equilibrium and heat addition/removal are minute differences based specifically on variables outlined in the delta-T calculation for the measurable time in question.

 

[toolmaker_03]

put simply if the loop is getting hot 1)add more rads or 2)add faster fans to the rads you have.

 

[Pschu]

Not being hostile (intentionally) but there are reasons the info exists; I've done it (first because I was uncertain if there really WAS any difference) and others have done it with actual test data to back it up.

Due to the high specific heat and thermal conductivity of water, there will always be much lower reported temps by hardware and your water delta will always be a function of flow rate, fans (airflow rate), radiator surface area and efficiency, ambient room temperature and total amount of heat in watts being introduced into the loop and removed at the same time.

Regardless of where you place water blocks, radiators or pumps, you are still moving water at the same rate, still adding the same amount of heat energy in watts and removing the same heat energy in watts. Differences of time to reach loop delta equilibrium and heat addition/removal are minute differences based specifically on variables outlined in the delta-T calculation for the measurable time in question.

Alright, this does make sense. Same amount of energy and cooling force both ways. So basically it will come down to my preference on how I order it. Thanks for your help and I apologize for my ignorance.

 

[rubix_1011]

My intention wasn't to imply ignorance at all- it was to suggest that the data in the sticky was there for a specific purpose; this topic has come up countless times before. Being new to watercooling does open up a Pandora's Box full of questions and uncertainties- just be patient, read, ask questions and ask more questions once you get those answers.

 

[Pschu]

Alright, sounds good. I am going to list off my blocks and pumps, any comments or concerns, feel free to let me know. I already did research on my own and feel pretty confident in what I am buying, but it doesn't hurt to check elsewhere.

Pump - Swiftech MCP655-B
Cpu block - XSPC Raystorm
Reservoir - XSPC dual bay
Gpu block - heatkiller GPU-X3
Rads XSPC rx120 and rx240
4 XSPC 2000 RPM fan on the 240 and 2 Coolermaster 2000RPM fans on the 120
also was gonna grab a danger den fillport to use as a drain off one of the lower Gpu block ports.
All of this is going inside a HAF 932 and cooling an i7 2600k and a GTX 670 FTW.

 

[rubix_1011]

Sounds like a pretty solid build. Any reason you went with the B-version instead of the Vario? Doesn't make much difference at all...just curious.

 

[Pschu]

I chose the B revision over the vario because I don't really think I would ever change the speed. Is the little bit of extra flow worth it from the vario? I know the B revision is locked at setting 4, and the vario can go to 5. I also saw there was a gain in flow between the two, but is it really worth it? Are there any other gains in buying the vario that I have overlooked?

 

[rubix_1011]

No, there isn't much that really would be a difference between them other than that. (Possibly price)

There are several sites that sell different versions of the same pump and for around the same cost as the Vario. Either way, both are very good pumps (since they are the same).

 

[Pschu]

I could get thee various model for 15 from where I'm ordering. Is it worth the extra money? There is also this pump http://www.frozencpu.com/products/13147/ex-pmp-131/Alphacool_VPP655_Variable_Speed_Pump_w_RPM_Monitoring_-_Single_Edition.html
What is the difference between that one and the MCP? Besides the fact that the one in the link is just the pump? Thanks.

 

[rubix_1011]

Same pump. MCP655, Alphacool 655, Koolance PMP 450 are all re-branded Laing D5's.

 

[Pschu]

So if I am just running the bay reservoir, would it really be worth it to pay the extra 15 and get the top and everything? Or save my money and get just the vario pump?

 

[rubix_1011]

Why not consider getting a bay/res to mount the pump? D5's are great pumps, but if you are considering that Alphacool one, you're going to need a way to mount it...you can't just run it as-is.

 

[Pschu]

I was gonna get the xspc dual bay one. Is that what you're talking about? I believe that should support it? http://www.frozencpu.com/products/11219/ex-res-232/XSPC_Dual_525_Bay_Reservoir_-_Laing_D5_MCP655_w_Blue_LED_Light_-_Clear_Silver_Black.html

 

[rubix_1011]

That's correct. If you go this route, the bay/res effectively becomes the 'top' for the pump.

 

[Pschu]

OK perfect. Is there anything bad to say about the bay reservoir? I like the idea so far. Thanks for all your help.

 

[rubix_1011]

There are a few guys around here that are using it and most seem to like it. It's the same config that you can get from XSPC as one of the pump upgrades for the Rasa and Raystorm kits.

 

[Pschu]

Best answer selected by Pschu.