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

Using Splitters.. Disadvantage?

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

curtisbouvier

Registered
Joined
Oct 16, 2006
Location
Canada
Just curious about what happenes when you use multiple splitters..

I want fresh water going to video card 1 and 2,

and I want fresh water going to CPU and RAM.

this would involve using 6 splitters,

Here is a diagram, should be more easier for you to understand.

http://img181.imageshack.us/img181/8095/test2copytj1.jpg

my question is simpley this, is there a downside to splitting your hoses like this?? do you start losing pressure, would water decide to have a higher flow rate down one tube with a weak rout down the other? or any type of negative situation?
 
First of all, you don't need to worry about supplying anything with "fresh" water. The flow rate in a typical loop is high enough that the coolant temp just doesn't rise that much in a single trip. You can use this formula to estimate the rise in temperature across any block:
°C = .0038 * Watts / gpm

And yes, there are some problems with running everything in parallel like that. The main problem is that you're dividing your flow among the components, hence each component gets less than it would in series. Sometimes, if you have a pump with low head and a high maximum flow rate (in other words, an aquarium pump), you can get better flow rates through restrictive components by putting them in parallel. But then you run into the other major downside: the least restrictive branch gets most of the flow, and usually the most restrictive component has to dissipate the most heat. Hence, balancing the flow rates in parallel branches can be quite a challenge.
 
It's not really worth the effort or pressure loss. You're only going to see a difference no greater (and usually less) of one degree between from when its "fresh" and after its gone through your waterblocks.
 
There is usually only .5c or less temp difference in the water across the loop. You will achieve better cooling going in series.

Also, cooling RAM will not achieve you any noticeable benefit as RAM puts out very little heat, and every RAM cooler commercially available seems to be 1/4" which kills flow in a standard 1/2" loop. A little bit of airflow is plenty enough for this.
 
EVen if your using a Peltier, one double radiator with at least 6000 BTU is enough ?
 
This doesn't seem to be related to using multiple wyes. Next time please start a new thread for a new topic. :)

khriez said:
EVen if your using a Peltier, one double radiator with at least 6000 BTU is enough ?
6000btu/hr = 1758 watts, which should be more than enough. But even at 100% thermal efficiency (which radiators do not achieve at high air flow rates), you'd need better than 200cfm actual flow through the rad with the coolant 15C over ambient. Considering that the air won't be heated to coolant temperature at that flow rate, your fans won't be able to drive as much air through the rad as they can move without restriction, and 15C over ambient is a lousy temp diff, 6000btu/hr is not very realistic. Try to find out what the difference between the coolant and air temperature was for that test and what sort of fan or cfm they used. Without knowing the test conditions, BTU ratings are completely meaningless.
 
Plus remember, water (like electricity) will take the path of least resistance. If your cooling your memory it would be custom to assume that the blocks have 1/4" fittings. Those will restrict the flow and create an imbalance in pressure. This will hinder temps. most likely.
 
striker85 said:
Plus remember, water (like electricity) will take the path of least resistance. If your cooling your memory it would be custom to assume that the blocks have 1/4" fittings. Those will restrict the flow and create an imbalance in pressure. This will hinder temps. most likely.


This is the KEY for new beginners who want to use parallel loops. ALWAYS use Y splitters b4 and after the SAME BLOCKS. Otherwise you alter the flow, one will has less flow than the other.
 
Split the flow and one loop will always dominate over the other. In the worst case, one loop might stagnate, producing little or no cooling, while the water takes the path of least resistance through the other loop.

Since flow rate basically controls cooling, splitting your flow in 1/2 will result in reduced flow and so reduced cooling.

If you want separate loops, then get separate pumps and rads.
 
orion456 said:
Split the flow and one loop will always dominate over the other. In the worst case, one loop might stagnate, producing little or no cooling, while the water takes the path of least resistance through the other loop.

Since flow rate basically controls cooling, splitting your flow in 1/2 will result in reduced flow and so reduced cooling.

If you want separate loops, then get separate pumps and rads.


Or if don right you end up with a slef balancing loop where the units that need less cooling get low flow and units needing more get the bulk of cooling flow ;)
 
Back