w/c setup rev1.1... with pic!

[arnoldma]

hi, made some changes to the last setup i posted, i took out ths chipset block, and changed ther fans from some 64cfm 26dba to some 125cfm 40dba ones with a fan controller!

the only thing im now not too sure about is the order of the componets, and the pump! i heard that the DD 12v pump is very loud?

also what is a good fan controller? cheers!

http://imagehost.biz/ims/pic.php?u=22395cjMB&i=31403&c=.gif

[Xenocide]

maybe im crazy but your waterflow direction is backwards if that tubing is going up from a DD pump...

[arnoldma]

yeah it is going up from the DD pump? do u think the pump cant handle it? :/

[Xenocide]

the pump pumps upward, and yes it can handle it

[arnoldma]

oh, so the one to the side is intake? and the one on the top is outtake? (i know, not a word)

[pwnt by pat]

Don't pump into the res and route your tubing so you use as little as possible and you will be golden.

[Xenocide]

yes your correct

the word your looking for is output

[greenman100]

you're movin the air the wrong way

xeno, you know better than to not catch that

[threeputt]

I think his rad is sitting outside the case...not pulling the air from inside the case but just above the top of the case...

[Alacritan]

The reservoir should be at the intake. You don't want to pump into your reservoir. Definitely not.

[pwnt by pat]

Here you go, rev1.2a:
http://img.photobucket.com/albums/v398/pwntbypat/3e84f725.bmp

[Alacritan]

Looks good, but if at all possible, put the radiator inside the case to minimize tubing and elevation change so your pump has to do less work (so you get better flow rates).

[MrKool]

The reservoir should be at the intake. You don't want to pump into your reservoir. Definitely not.


why not? i got mine setup so that the water flows from one side of the reservior to the radiator, which then colds the water and send it to the CPU block then to the pump and pump then puts the hot water from the cpu back into the reservior. My idle is 34 and load is 42. My P4 3.0C is running at 4.0 Ghz with 1.6 volts so that is why my temps are little high. at 3.7 my idle is 29 and load is 38.

Also i got my Rad on the outside of the case as well :) i have a black ice extreme 2 and i could not fit it insde my antec soho case.

[dj_2004]

I will have the same exact setup just about, cept pump will be by the front of the case and will use a single 120mm rad. Is pwnt by pat's setup plan good? I thought it goes pump-rad-cpu-vid-res-pump. Thx!

[Lancelot]

Still the ideal situation is when your pump sucks in from the reservoir, then to the radator and then the cpu block...

[greenman100]

why not? i got mine setup so that the water flows from one side of the reservior to the radiator, which then colds the water and send it to the CPU block then to the pump and pump then puts the hot water from the cpu back into the reservior. My idle is 34 and load is 42. My P4 3.0C is running at 4.0 Ghz with 1.6 volts so that is why my temps are little high. at 3.7 my idle is 29 and load is 38.

Also i got my Rad on the outside of the case as well :) i have a black ice extreme 2 and i could not fit it insde my antec soho case.


http://www.overclockers.com/articles1088/

order makes no difference in temps, and there is no such thing as "hot" and "cold" water, as the temp caries by 1C or less

he made the commnt on the radiator because centrifugal pumps work better if they have an unrestricted inlet

[thorilan]

real world results prove that you can get more than 2C diference by setting up your components diferently

like putting the pump before the radiator and putting the res before the pump

[pwnt by pat]

Resevious in NO WAY cool the water (unless you have the reserator). There is simply not enough surface area or time for the water to release the heat before it's sent through the loop again. Also, in a closed system, any air traped in the res can only absorb so much heat until it reaches an equilibrium with water temperature.

Having a unrestricted pump inlet is why you should (almost) never have componants between a res and the pump.

The only time when you might see a difference in temperatures when having the pump before the cpu block is when you are using huge pumps in excess of several hundred watts. Even then it's not that bad. True story: Right now my friend as a 700w 1200gph 22.5ft head SUMP pump submersed in an aquarium. It has a completely unrestrected inlet so its efficiency is greater than if it was modded to be inline. There is no radiator in this setup, it's completely passive and silent. His a64 3000 idles at a solid 40c.

[arnoldma]

lol, thanks guys for the reply, wasn't expecting so many :)

anyway, i'm a bit confused now, some of you are saying what i have is ok, to have the rad. before the cpu block, but other are saying pump before the block, i.e. the rev 1.2 diagram drawing by "pwnt by pat" (tahnks btw)

the only thing i thought was if i have the pump before the cpu block wont the pump warm up the water a little? but then again i see how the pump should have unrestricted flow, i.e. sucking water from the res.

decisions decisions :P

ps. i got a 3.0e D0 with abit ic7-max3

[Xenocide]

there is no such thing as warmer water blah blah blah dont' worry about it

[pwnt by pat]

The unrestrectided inlet on the pump will allow it to have greater effeciency and lack of pump head loss compaired to having it last in the loop. The benefits of this will outweigh any performance benefits of rearanging the other componants so its rad>cpu (which should be no more than 1c anyway, probably closer to .5c)

[greenman100]

the only thing i thought was if i have the pump before the cpu block wont the pump warm up the water a little? but then again i see how the pump should have unrestricted flow, i.e. sucking water from the res.



did you bother to read my link?

[arnoldma]

sorry.. just read it now, i guess ill be taken pwnt by pat as rev1.2 and start working on 1.3 :D thanks a lot guys!