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

most efficient liquid

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

striker85

Member
Joined
Mar 7, 2006
Location
Connecticut
Ok, there are many arguments about what to use for the liquid in liquid cooling. When I had my first setup, it was a kit that came with a water/antifreeze mix. But the instructions said to use distilled water. I put that in and behold, a few weeks later it rusted up and seized. So now I have my new parts coming in for my system which I vowed will be done the right way. So far I have this:

pump: pond mag2 250gph, 12' well head
cpu block: swiftech apogee
gpu block: zalman zm-gwb1
tubing: just some home depot 1/8" thick 3/8" ID clear plastic stuff
radiator: a 1976 corvette heater core, this thing is perfect 6.25"x6.25"x2.5" with 3/8" and 5/8" barbs(which will have to be modded a little, of course)


Now, my cpu block and gpu block are copper and aluminum. I know there can be a problem there, but the gpu block is anodized, I'm pretty sure. I know that to lubricate the pump, though the mag2 says it doesn't need lubrication, and to prevent corrosion and bacteria growth there should be something like 10% antifreeze in there. But I have read threads about antifreeze not being enough. So should I just use a mixture of 90% distilled water with 10% antifreeze with a few drops of iodine? What would provide the safest/most efficient solution?
 
Before I even answer your question I just want to say, lets try the search feature next time. This is seriously the 10th thread this week ive seen on which liquid is the best. Sorry if I sound rude, I dont mean to be, its just repetitive.

Just go with the 90% distilled water, and 10% anti-freeze. Im sure all those other liquids you can buy are great but why get them when you can just do it yourself for cheaper?
 
Trust me, I've read just about every thread on cooling additives. But so many answers to what seems like the same question, but certain details are different, just get confusing. I just wanted to make sure about what I had read.
 
It actually rusted????
Its weird, rust in submerged applications take a long time because rusting needs constant air.
It might have corroded which is a different thing.
Usually 10% antifreeze will suffice unless you have un-anodized alum components in which case you are going to need to up the antifreeze/water ratio to 15% or in extreme cases 20%.
 
If you're going to mix copper and aluminum, I'd recommend 10% Zerex racing coolant. This is the anti-corrosive stuff found in antifreeze, without the antifreeze itself. At 10%, it keeps mixed-metal automotive cooling systems at from corroding at 80C, hence you shouldn't have too much trouble. Of course, all things being equal, sticking with copper, brass, and plastic eliminates a potential headache and gives you better heat transfer as well.

(Edit: There is also Zerex antifreeze, but it's a completely different product. What I said above applies only to the racing coolant. If you use antifreeze, you'll need at least 15% as Voigts recommends below.)

Distilled water, btw, is the most efficient cooling liquid for our purposes. The higher the ratio of water to other stuff, the better it will perform. But besides causing leaks, corrosion will make your parts less efficient at transferring heat, so be conservative and use plenty of corrosion inhibitors.
 
Last edited:
It would be adviseable to go with the zerex or another type of antifreeze in at least a 15% to 85% distilled water mixture due to the mixed metals. Also, to make sure you have no growth, use some fish tank type algaecide and/or a few drops of non-alcohol base iodine like walmart "Equate povidone-iodine 10%". I use both of those and have no growth whatsoever. Some people can use the antifreeze alone and not have any trouble, however others do and so it is best to do the algaecide and/or iodine.
 
mixing metals does not need more than 10%. in a large capacity system it needs even less.

for a 4liter plus system you can go as low as 3% easily( but if oyu have a 4 liter system odds are you arnt worried about the minor diference anyways)
 
Really? Maybe mixing aluminum and copper is less of a problem than I thought. How many systems like that have you built and looked at again after a couple years?
 
Otter said:
Really? Maybe mixing aluminum and copper is less of a problem than I thought. How many systems like that have you built and looked at again after a couple years?

Said it tens of times I had a loop with bare Al and bare Cu for a year with a 80/20 mix with no problems. The Al looked brand new when I drained the loop. It was my HTPC/downloading box and ran 24/7
 
To re-affirm what striker mentioned, there's some clashing of ideas on this subject. But, there are some very notible similarities.

When all my WC'n parts come in, I'm going to use Fluid XP+. But that's just me, I'm willing to pay extra for realiability even if it reduces effeciency (slightly).
 
TreeNode said:
To re-affirm what striker mentioned, there's some clashing of ideas on this subject. But, there are some very notible similarities.

When all my WC'n parts come in, I'm going to use Fluid XP+. But that's just me, I'm willing to pay extra for realiability even if it reduces effeciency (slightly).


::starts digging through threads to find the one about fluid xp becoming conductive over time::
 
jamesavery22 said:
::starts digging through threads to find the one about fluid xp becoming conductive over time::

I don't think he cares about conductivity. He seems more worried about corrosion.
 
HAh... I saw this video of some guys pouring FluidXP on top of some some computer parts while they were on and it worked fine, then they poured it directly on (and into) the PSU and for about 5 seconds it worked then BANG!

Funny thing is, it turned on again and they kept doing it till it died.

So basically I'd rather have something that just might possibly save my computer from turning to dust if it leaked versus something that simply won't, plus, less corrosive :)
 
jamesavery22 said:
Said it tens of times I had a loop with bare Al and bare Cu for a year with a 80/20 mix with no problems. The Al looked brand new when I drained the loop. It was my HTPC/downloading box and ran 24/7
And yet, I've seen some very sad pictures of corroded blocks. Still, I think I should tone down my paranoia about mixed metal loops. I don't want to propagate misinformation. :)
 
Back