Dual Loop designs.

Tachyon · 08-08-2004, 11:19 PM

Greets one and all. New to the forums, veteran with the puters.
I've been working on ideas (in my head) for years now about different things. And it's about time I make some of them a reality. Some of these ideas have under gone several changes since technology has changed. I've always thought that an isolated, cool supply of coolant for the CPU, GPU, NB would optimal. I've seen mentioned here "dual loop" systems. Knowing about the "path of least resistance" are the dual loop designs a single pump with a manifold design or a dual pump, totally isolated variety? My idea is to have dual pumps (or dual impeller/single motor) fed from dual supplies lines from a common reservoir. One cooling the CPU/NB. The other going to the GPU. Both either rejoining in a common radiator or a 2nd service reservoir then the radiator. The “supply” reservoir being a "sealed" aquatube, the service reservoir being an externally rear mounted one for filling and bleeding.

ps Tim Allen suggested this aquarium pump. Will it be ok?

SlaterSpeed · 08-09-2004, 04:20 PM

Thats a bit like what i have. I have 2 seperate loops one uses 1/2"id tube and cools the cpu only where as the other cools everything else (nb,sb,mosfet,gfx etc.) and is 8mmid. They both have a seperate pump and rad but they share the same res. The idea being the cpu gets a nice highflowrate as theres not added resistance by other blocks.

The easyest way to do it is to go for totaly seperate systems. Im not 100% sure how using a common res will attualy effect matters. The 2 loops would share coolent and that would mean that the total heat energy would be 'pooled' so it would be more like running the differant componants in paralell than in a tyotaly differant loop. Cant see any problems there myself.

slater..

bigben2k · 08-09-2004, 06:18 PM

Welcome to ProCooling!

A dual loop really isn't necessary, but it's certainly an option. The temperature variation within a single loop is actually very small (i.e. within a fraction of a degree celsius), unless you have an unusually low flow rate.

BTW, the tim Allen pump would hardly qualify...

Here's one that's really overkill:
http://www.gouldspumps.com/cat_pumps...atid=76&step=4

tex707 · 08-10-2004, 12:28 AM

I've got a dual loop system, but not one of a kind you are referring to. There is an inner (closed) loop and outer (open) loop. Closed loop is a classic watercooling setup, except for the fact that the radiator is not cooled by air but by water from the outer loop, which is actually water that circulates inside the bong and a large thermally insulated reservoir.

msv · 08-10-2004, 01:18 AM

Two and a half metric ton of pump...
Should be enough even for dual loops. My Eheim 1048 wasn´t really up to it. Should have tried one of those, Ben.
regards
Mikael S.

Tachyon · 08-12-2004, 04:13 PM

Did he say bong ??

bigben2k .... thanks and .... gaud I'd be worried that thing would suck my whole computer right in!!

I'd like to design a system that does the most efficient cooling it can w/o any exotic measures. There are a few things that I have been considering that are a bit out of the "norm". Like an aquarium chiller to bring coolant temp to ~40`F etc. But until I win the lotto I'd like to try the simple stuff first. Supplying each subsystem it's own "cool" supply of coolant w/o too much extra hardware seems possible. The pooling of the thermal energy in a common reservoir shouldn't be a real issue as it's not an issue in a non-dual loop system. I'm just trying to get around sending the CPU's heated coolant on to the GPU which now days creates enough of it's own thermal output.

p.s. anyone ever see a dual impeller/single armature pump?

08-08-2004, 11:19 PM	#1
Tachyon Cooling Neophyte Join Date: Aug 2004 Location: 41.39x88.31 Posts: 11	Dual Loop designs. Greets one and all. New to the forums, veteran with the puters. I've been working on ideas (in my head) for years now about different things. And it's about time I make some of them a reality. Some of these ideas have under gone several changes since technology has changed. I've always thought that an isolated, cool supply of coolant for the CPU, GPU, NB would optimal. I've seen mentioned here "dual loop" systems. Knowing about the "path of least resistance" are the dual loop designs a single pump with a manifold design or a dual pump, totally isolated variety? My idea is to have dual pumps (or dual impeller/single motor) fed from dual supplies lines from a common reservoir. One cooling the CPU/NB. The other going to the GPU. Both either rejoining in a common radiator or a 2nd service reservoir then the radiator. The “supply” reservoir being a "sealed" aquatube, the service reservoir being an externally rear mounted one for filling and bleeding. ps Tim Allen suggested this aquarium pump. Will it be ok? __________________ Tach - The two most common elements in the universe are hydrogen and stupidity! Last edited by Tachyon; 08-08-2004 at 11:58 PM.

08-09-2004, 04:20 PM	#2
SlaterSpeed Cooling Savant Join Date: Nov 2003 Location: Suffolk, UK Posts: 234	Thats a bit like what i have. I have 2 seperate loops one uses 1/2"id tube and cools the cpu only where as the other cools everything else (nb,sb,mosfet,gfx etc.) and is 8mmid. They both have a seperate pump and rad but they share the same res. The idea being the cpu gets a nice highflowrate as theres not added resistance by other blocks. The easyest way to do it is to go for totaly seperate systems. Im not 100% sure how using a common res will attualy effect matters. The 2 loops would share coolent and that would mean that the total heat energy would be 'pooled' so it would be more like running the differant componants in paralell than in a tyotaly differant loop. Cant see any problems there myself. slater.. __________________ aka. slater3333uk - The self proclaimed 'Middle Player' 'Liquified'

08-09-2004, 06:18 PM	#3
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Welcome to ProCooling! A dual loop really isn't necessary, but it's certainly an option. The temperature variation within a single loop is actually very small (i.e. within a fraction of a degree celsius), unless you have an unusually low flow rate. BTW, the tim Allen pump would hardly qualify... Here's one that's really overkill: http://www.gouldspumps.com/cat_pumps...atid=76&step=4 __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

08-10-2004, 01:18 AM	#5
msv Cooling Savant Join Date: Dec 2002 Location: Sweden Posts: 336	Two and a half metric ton of pump... Should be enough even for dual loops. My Eheim 1048 wasn´t really up to it. Should have tried one of those, Ben. regards Mikael S. __________________ The only constant factor in all Your failures is You. Catapultam habeo. Nisi pecuniam omnen mihi habis, ad caput tuum saxum immane mittam.

08-12-2004, 04:13 PM	#6
Tachyon Cooling Neophyte Join Date: Aug 2004 Location: 41.39x88.31 Posts: 11	Did he say bong ?? bigben2k .... thanks and .... gaud I'd be worried that thing would suck my whole computer right in!! I'd like to design a system that does the most efficient cooling it can w/o any exotic measures. There are a few things that I have been considering that are a bit out of the "norm". Like an aquarium chiller to bring coolant temp to ~40`F etc. But until I win the lotto I'd like to try the simple stuff first. Supplying each subsystem it's own "cool" supply of coolant w/o too much extra hardware seems possible. The pooling of the thermal energy in a common reservoir shouldn't be a real issue as it's not an issue in a non-dual loop system. I'm just trying to get around sending the CPU's heated coolant on to the GPU which now days creates enough of it's own thermal output. p.s. anyone ever see a dual impeller/single armature pump? __________________ Tach - The two most common elements in the universe are hydrogen and stupidity! Last edited by Tachyon; 08-12-2004 at 05:29 PM.

08-10-2004, 12:28 AM	#4
tex707 Cooling Savant Join Date: Apr 2001 Location: Europe Posts: 164	I've got a dual loop system, but not one of a kind you are referring to. There is an inner (closed) loop and outer (open) loop. Closed loop is a classic watercooling setup, except for the fact that the radiator is not cooled by air but by water from the outer loop, which is actually water that circulates inside the bong and a large thermally insulated reservoir.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)