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

Choosing+building a CPU Water Block

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

sirbow2

Registered
Joined
Jun 13, 2010
I have researched a lot about DIY Water Cooling. i have come up with many water block that i could easily make but dont know which one is the best :confused: .

i will just post some pics of the ones i was thinking about and you could help me pick the best one. :)

Option 1: 1/4in copper tubes soldered together, then flattened on the bottom, then maybe a thin copper plate soldered to the bottom of the tubes, or just the flat bottom Tubes directly touching the CPU

Similar Example:
img0857zo8.jpg


option 2: a 3/8in or 1/2in thick 2x2in copper block with 1/4 holes (maybe 5 or 6 holes) drilled through with 1/4 copper pipe connecting with the holes

Similar Example:
c7a4c63980.jpg


option 3: a piece of 1/8 in copper plate with a acrylic cap with hose Barbs

Similar Example: NOTE: the block in this pic is in the bucket, best pic i could find :(
4caff97ecf.jpg


option 4: a 1/4in piece of copper with a maze path 1/8 in deep in to the copper, acrylic cap with barbs, similar to option 3


those aren't my blocks i just used them to help explain

thanks :)
 
OP
S

sirbow2

Registered
Joined
Jun 13, 2010
ok thanks for the advice for whenever i make my block i just wish i needed water cooling so i had a excuse to make this :D

Conumdrum, when you said "Seen this done somewhere else in another of the 15th variant." what did you mean? typo?

My air cooling setup:
even when my fans are turned down all the way i get really good low temps, 38C GFX, 28C CPU, this is when it is about 85F in the room too! all my fans are controlled by a FanMaster controller, enzotech Xtreme CPU cooller, stock GFX cooler, and 3- 80cm case fans

EDIT\\\ Also i have heard than if i have a T- coupling in my loop, i dont need a reservoir, if so where does the T-coupler need to be located? i was thinking the top of the case
 
Last edited:

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
EDIT\\\ Also i have heard than if i have a T- coupling in my loop, i dont need a reservoir, if so where does the T-coupler need to be located? i was thinking the top of the case
You put the T fitting in-line just in front of the pump inlet. The open end of the T can go anywhere, really, depending on how you have your loop routed. I've got one T attached to a fillport that's installed in the top of the case. In another loop the T is external and runs up the outside of my case. Both T's are about 1' long and are over-sized to 3/4" tubing ... :)
 

m0r7if3r

Member, Water Cooling Sticky Reading Enforcement O
Joined
Jun 17, 2010
Location
Marietta, GA
ok thanks for the advice for whenever i make my block i just wish i needed water cooling so i had a excuse to make this :D

Conumdrum, when you said "Seen this done somewhere else in another of the 15th variant." what did you mean? typo?

My air cooling setup:
even when my fans are turned down all the way i get really good low temps, 38C GFX, 28C CPU, this is when it is about 85F in the room too! all my fans are controlled by a FanMaster controller, enzotech Xtreme CPU cooller, stock GFX cooler, and 3- 80cm case fans

EDIT\\\ Also i have heard than if i have a T- coupling in my loop, i dont need a reservoir, if so where does the T-coupler need to be located? i was thinking the top of the case

just be aware that it will take a lot more time to fill your loop and might take a bit longer to bleed the bubbles. for a novice reses are easier though.
 

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
It does take longer to fill a loop with a T-line, even one of mine, but I'd bet my T's bleed as fast as a res. I use a 1/2-3/4-1/2" T and 3/4" tubing for the T-line. At a foot long it holds about half what a MicroRes holds and few bubbles can jump that 3/4" gap. It also helps to be creative with the mounting. Having the intake coming in slightly "uphill" helps ... ;)
 

m0r7if3r

Member, Water Cooling Sticky Reading Enforcement O
Joined
Jun 17, 2010
Location
Marietta, GA
It does take longer to fill a loop with a T-line, even one of mine, but I'd bet my T's bleed as fast as a res. I use a 1/2-3/4-1/2" T and 3/4" tubing for the T-line. At a foot long it holds about half what a MicroRes holds and few bubbles can jump that 3/4" gap. It also helps to be creative with the mounting. Having the intake coming in slightly "uphill" helps ... ;)

:thup: this is the typical experience, but I have heard of some people saying bubbles take longer to bleed...figured it was worth at least mentioning
 

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
:thup: this is the typical experience, but I have heard of some people saying bubbles take longer to bleed...figured it was worth at least mentioning
With a "standard" 1/2" barbed T that may be true. My T's are copper plumbing T's - no barbs required and no added restriction, either. For some things off-the-shelf just doesn't cut it ... ;)
 

m0r7if3r

Member, Water Cooling Sticky Reading Enforcement O
Joined
Jun 17, 2010
Location
Marietta, GA
With a "standard" 1/2" barbed T that may be true. My T's are copper plumbing T's - no barbs required and no added restriction, either. For some things off-the-shelf just doesn't cut it ... ;)

you and your crazy parts :D
 
OP
S

sirbow2

Registered
Joined
Jun 13, 2010
thanks for all the info guys but ive read around a little bit and i found that the Phenom II X6 1055T has a x14 Multi but in bios i can go to x16.5. i have a 1055T?!
 

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
thanks for all the info guys but ive read around a little bit and i found that the Phenom II X6 1055T has a x14 Multi but in bios i can go to x16.5. i have a 1055T?!
Are you sure it will actually go higher than 14X? Have you checked it with CPU-Z? Seems I remember someone else saying the same thing then found out that even though the BIOS looked like that it didn't really happen ...
 
OP
S

sirbow2

Registered
Joined
Jun 13, 2010
yep your right i was just thinking that and its true set the multi to x16.5 but it still only goes to 200x14 in prime95, darn i thought i was lucky :(
 
Last edited:
OP
S

sirbow2

Registered
Joined
Jun 13, 2010
i easily got to 3.8ghz by 14x270 that was stable but too hot for my tastes, about 60C full load, no volt increase. so i lowered it to 3.5ghz which is about 40C full load, im gunna see how low i can get the voltage with 3.5ghz

now i may have an excuse to build that water block :D :)

also has anyone have issues with a 1055T w/ GA-890GPA-ud3h reporting wrong temps? becasue core temp + cpu-z have said the CPU was 18C(64F) in a room that was more like 26C(80F), this is while all my case fans were on low, maybe off by 10C?
 
OP
S

sirbow2

Registered
Joined
Jun 13, 2010
this is probably a real noobish question but what do people mean when they say folding, like on a CPU, is it the same as OCing
 

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
Folding ([email protected], aka [email protected]) and crunching ([email protected] or Rosetta) are Distributed Computing (DC) projects that Overclockers.com participates in as teams. Crunching rigs can either be run only when your machine is idle or you can run it all the time, even when the machine is in use. (I can't remember if [email protected] has a "Use only if idle" option or not.) Either way, all those projects stress the CPU as much or almost as much as Prime95 and, because of the extended time processing a work unit and the large numbers involved, actually stresses the CPU and RAM more than most benchmarking programs. Both projects can also use the GPU (video card) to process work units ...
 
OP
S

sirbow2

Registered
Joined
Jun 13, 2010
but whats the point of doing it, whos has the best pc or something?
 

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
All DC projects are about donating your computer time to the advancement of science. It's kind of a social thing, too - we compete with other teams once or twice a year and egg each other on to do better but in the end it's about donating the computer time. We all put so much time, effort, and money into revving up our rigs that many of us feel it's kind of a waste to let it just sit there on the desk 20 hours a day doing nothing ...