Pro/Forums

Pro/Forums (http://forums.procooling.com/vbb/index.php)
-   General Liquid/Water Cooling Discussion (http://forums.procooling.com/vbb/forumdisplay.php?f=9)
-   -   New to H2O..a few setup questions (http://forums.procooling.com/vbb/showthread.php?t=3812)

Jimbo 07-19-2002 07:45 AM

New to H2O..a few setup questions
 
I've finally gotten together most of what I need for my first water setup. I purchased a Danger Den chipset block to go along with a Maze 3 but I'm not sure of the best way to setup these two blocks.

The pump I currently have is an 600LPH/158GPH Eheim 1048 and I went with a cheaper Black Ice Prime radiator. To get to the point. I would like to setup both the cpu and chipset blocks but I'm concerned about too much heat being delivered to the Chipset block from the cpu if I run the two in series.

If I go with a Y splitter at either side of the two blocks, is the 1048 enough to handle two splits in the tubing? Can anyone direct me to some good links which break down the amount of force needed to handle various tubing setups? Thanks in advance.

-jim (hot stank) M.

hmale 07-19-2002 08:39 AM

There are different views concerning the setup of multiple blocks within a closed circuit. Some favor using a "Y" to split the water flow through the blocks (parallel configuration). Others (me included) prefer a series configuration where the flow is not spit and the velocity of the water through the blocks is maximized.

In practice I believe that there is validity in both schools of thought. I am a firm believer in testing results yusing various configurations and going with what is best for your particular circumstances. That is what I do with all my projects.

There is no one set answer as there are so many variables that impact cooling e.g. back pressure, component placement, number of elbows and tees, pump capacity, radiator design, radiator fan selection, case cooling, and many others.

The pump you intend on using is tried and proven and should be sufficient for what you are planning.

To be honest, I do not believe that the difference between the parallel Vs series circuit for the components you propose is going to impact temperatures more than +/- 2 degrees C.

Best of luck in your journey. Half the fun is getting there!!

gmat 07-19-2002 08:45 AM

Wow it's a tougher question that one would think.
Actually 600 l/h is a lot already. In your case i dont think putting the blocks in parallel (with 2 Y's) would yield far better results than putting them in series. Actually water temps dont increase that much (generally no more than 1 or 2°C), so the NB wont fry. So go the easy way, Pump->rad->cpu->NB->pump all in series, make sure you've got a minimum of turns and you'll be OK.
On the other hand if splitting with a Y give you *less* turns (ie no need for elbow fittings for example) go that way (parallel):

Code:

pump->rad->(Y)->CPU->(Y)->pump
              \->NB->/

The major flow killer is turns and elbows.
In short, do what's easier and cleaner for you. It will reflect magically on the performance.

Ah the 1048 is quite enough for most 'usual' applications. Just glue that impeller :p

(edit) Hehe hmale beat me to it :p

decodeddiesel 07-19-2002 10:58 AM

In this case I am deffiniatly thinking you should put them in series. As was stated before your watertemp is really only going to rise 1 or 2C comming through the CPU block, and the NB really doesn't have to be kept at THAT low of a temp, even with 25C water flowing through the block your bridge has such a large surface area it will still be kept very cool. I am afraid if you start adding Ys and 90s to you loop you will have too great an impact on flow and you may cause more harm than good. Hope this helps.

bigben2k 07-19-2002 11:11 AM

Well, I agree, the blocks in series is very simple, and the pump is sufficient, but...

You could use T type fitting to feed the NB wb, in parallel. Even more, add a valve to the NB WB loop. Here's why:

The NB WB doesn't generate anywhere near as much heat as the CPU does, so you'd want a relatively proportional amount of flow. The advantage of course, is that the CPU WB gets most of the flow, and that's what you really want.

Alternatively, run the WBs in parallel, but feed the NB WB with 1/4 ID tubing, from a T fitting (instead of a valve).

and that's my $.02 !

Jimbo 07-19-2002 11:34 AM

Wow..thanks for all the input. I'm going to test everything out with the two blocks in series and then go from there. Waiting for a few last parts from the UPS man today and then I can get going. A few additional points and questions...

After reading through the very helpfull and thorough H2O Technologies Explained in the articles section, I'm going to add an air trap made from a sealed electrical box. Regarding the air trap...should the water level be above the entrance (upper, highest point) barb or below it?

Thanks again for all your assistance...I'll keep everyone posted on the outcome...specs for the project:

-Epox 8KHA+
-Unlocked T-Bird 1.33 (to start)
-256Meg Nanya memory
-PNY Ti4400 w/ Blorb and ramsinks
-Maxtor 20Gig
-Lite-On 16x DVD
-Maze 3/DD Chipset block
-Black Ice Prime rad.
-(2) 120MM fans w/ homemade fan controller
-Eheim 1048

bigben2k 07-19-2002 11:53 AM

Quote:

Originally posted by Jimbo
Regarding the air trap...should the water level be above the entrance (upper, highest point) barb or below it?

It's best if the waterline is above, because it limits the water from splattering everywhere, making bubbles/foam and noise.

decodeddiesel 07-19-2002 01:04 PM

Most deffiniatly agreed, I have made the mistake of keeping an air trap so the water "splashed" in it. Made a lot of unneeded noise and foam.


All times are GMT -5. The time now is 05:14 AM.

Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
(C) 2005 ProCooling.com
If we in some way offend you, insult you or your people, screw your mom, beat up your dad, or poop on your porch... we're sorry... we were probably really drunk...
Oh and dont steal our content bitches! Don't give us a reason to pee in your open car window this summer...