Two radiators in a single loop?

I already have a single 120 mm Black Ice rad for my A64.
I'm looking to add a chipset block and a block to my pcie 6800GT.
would adding another rad be a good idea?

or what should I do?

Thanks in advance for suggestions

Brian

I don't see any harm in adding a second rad. I currently have two rads in my system, a HC and a BIX cooling a pair of 6800GTs and my A64 3500+ Clawhammer.

I would keep the rads in series and put your second rad between two components. For example for your setup this would probably be best:

pump -> rad1 -> CPU -> rad2 -> GFX -> NB -> pump

The first rad is a 1x120 Black Ice Pro
I was thinking of adding a 2x120 Black Ice Pro to blowholes that I would cut in the top of my Antec P160 case since there's no room elsewhere.

Also, pump problems?
here it is


thanks for the suggestion

Your pump is fine. I would add only the GPU block and see what that does for your temps, overclock, and stability. You may find the results are still acceptable without adding more rad capacity. The CPU temp may go up a few degrees, but if it's still stable, is that good enough for you? If you do add another rad to the loop, series is probably the best although it wouldn't hurt to try just the rads in parallel.

Is there a history of better overclocking when actively cooling the NB on your board? If not, I'd just put a passive heatsink on the NB.

The order of the components in a serial water loop does not matter for practical watercooling purposes, just make the tubing runs short with gentle bends.

I just have whatever the cheap UV tubing is off of dangerden, and it doesn't bend very well. I will be getting Tygon, is this a mistake? Will Clearflex do?

My case is a mid-tower, so it's quite small for all this extra stuff, and doesn't have a lot of room to bend.

Enigma422 said:

I don't see any harm in adding a second rad. I currently have two rads in my system, a HC and a BIX cooling a pair of 6800GTs and my A64 3500+ Clawhammer.

I would keep the rads in series and put your second rad between two components. For example for your setup this would probably be best:

pump -> rad1 -> CPU -> rad2 -> GFX -> NB -> pump

Click to expand...

that ^^ would probably be the best solution, b/c the CPU will heat the liquid then go into another cooler be cooled and then get some more heat and then go into a cooler again.

very good idea if i do say so myself.

bmartin said:

I just have whatever the cheap UV tubing is off of dangerden, and it doesn't bend very well. I will be getting Tygon, is this a mistake? Will Clearflex do?

My case is a mid-tower, so it's quite small for all this extra stuff, and doesn't have a lot of room to bend.

Click to expand...

i had tygon before make sure you get a thickwalled stuff, it'll be less bendable.

honhon said:

that ^^ would probably be the best solution, b/c the CPU will heat the liquid then go into another cooler be cooled and then get some more heat and then go into a cooler again.

very good idea if i do say so myself.

Click to expand...

The order does not matter and it does not work like you are describing above. The water temp does not change quickly due to its large thermal capacity and it certainly doesn't bounce up and down 5-10C as it goes around the water loop. Do you really think it changes that much in one pass through a block or radiator? The way radiators work is not intuitive but they are key to understanding how watercooling works.

What actually happens is the water and cpu temps slowly rise as the blocks add a steady amount of heat to the circulating water. As the water temp rises, the radiator starts to dissipate heat more effectively. The water temp will continue to rise until equalibrium occurs between the heat added by the blocks and heat dissipated in the rad. The heat dissipated by a rad is controlled mainly by the temperature delta between the air temp and water temp.

A given rad can dissipate much more heat IF you allow/accept the delta between the air and water temps to rise. Increased airflow across the rad may be required for the higher heat dissipation.

The cpu temp is determined by the temp of the water in the loop plus the C/W rating of the water block multiplied by the CPU power dissipation.

gungeek said:

The order does not matter and it does not work like you are describing above. The water temp does not change quickly due to its large thermal capacity and it certainly doesn't bounce up and down 5-10C as it goes around the water loop. Do you really think it changes that much in one pass through a block or radiator? The way radiators work is not intuitive but they are key to understanding how watercooling works.

What actually happens is the water and cpu temps slowly rise as the blocks add a steady amount of heat to the circulating water. As the water temp rises, the radiator starts to dissipate heat more effectively. The water temp will continue to rise until equalibrium occurs between the heat added by the blocks and heat dissipated in the rad. The heat dissipated by a rad is controlled mainly by the temperature delta between the air temp and water temp.

A given rad can dissipate much more heat IF you allow the delta between the air and water temps to rise. Increased airflow may be required for the higher heat dissipation.

The cpu temp is determined by the temp of the water in the loop plus the C/W rating of the water block multiplied by the CPU power dissipation.

Click to expand...

very interesting. So lets say i put a temperature probe in the water just after the cpu, and right after the rad... so there wouldn't be a big difference in temps?

lowfat said:

very interesting. So lets say i put a temperature probe in the water just after the cpu, and right after the rad... so there wouldn't be a big difference in temps?

Click to expand...

Less than 1C water temp delta around the loop.

i ha 2 rads and the last rad was restrictive and sucked hot air from the back of my case and it was still 1.5 cooler then just one rad. if i could fit another bix and draw cool air in it would help 3C with the laing/dangerden ddc. I modded mine with a 3/8 barb on top and did not see any improvement as flow is not my weak link 1 rad and 80cfm is my weak link now.

That's my other question, how much difference would say, a 50cfm fan make on a 1x120 rad as opposed to a 90cfm fan?

Edit: I am upgrading right now and am waiting for my new mobo and psu, so I was also rethinking my water cooling setup, and it occured to me that I have the 50cfm Antec case fan that was included with my case on the radiator. I also have a considerably louder 90cfm sunon fan that could be put on the radiator. I can't test that now, but it should make quite a difference, right?

I don't think a rad fan has to be loud to be effective. The only way to really know is try the 2 fans and see how much better the loud one works.

Personally, I'd give up 1 or 2C in cpu temps to have a quiet fan (and system). I overclock (but not hardcore about it, no volt-mods for me) and there are 120x38mm fans that are not loud but push a decent amount of air.

Using a 120x38mm fan will have a noticable advantage over the thinner 120x25mm fans. The deeper fans produce more pressure which helps move more air through the (relatively) restrictive heater core fins. You could always put one on a fan controller or wire it direct for 7V or 5V. I have a 140 CFM fan running at 5V and it moves enough air and is much quieter than operating it at 12V.

I have two censors measuring the water temp before and after the rad. Right now, before is 24.9 and after 24.1.

mulri rads

bmartin said:

I already have a single 120 mm Black Ice rad for my A64.
I'm looking to add a chipset block and a block to my pcie 6800GT.
would adding another rad be a good idea?

or what should I do?

Thanks in advance for suggestions

Brian

Click to expand...

888888888888888

This may sound odd but I really have four rads inline from off the block then to the reservior. It works. I have a large pump outside the case. I tried it the other way around and the results were the same for me. In the case---is that what you want{?}---i dunno what problems you might have.

With four rads I increased the water quantity about a pint. It is good to have a base volume of water that is large.

After hours of running the PC the water temperature was about 5F over the ambient temperature. Basically, I feel no heat at all with my hand on the tubes coming or going, but the overall heat of the water increases a bit. I have over a gallon total. This. of course, is external to the case.