Modding What is the best water-cooling coolant?

We take a look at four of the latest non-colour coolants to see which is the best

http://www.bit-tech.net/hardware/cooling/2012/03/26/what-is-the-best-water-cooling-coolant/1

 

Can we have a comparison of dielectrics sometime? Preferably including the (probably) rather expensive Fluorinert coolant?

 

Here's what should be done to compare..

Head to the petrol station. Buy a) battery water (distilled water) and b) antifreeze. Mix the two together, perhaps at different ratios for the sake of comparison.

It'll be a hell of a lot cheaper than all the products in this test, and I'll bet my life it'll perform just as well, if not better.

 

Still no feser one!

 

I don't find that surprising, personally. I'm sick of that crap.

Then again, this type of test probably wouldn't pick up Feser One's primary fault - Olympic medal-winning gunking, after sustained, long term use.


BTW Antony, I thought the Mayhems Aurora fluids were not designed for 24/7 setups, and more for show systems? Good news if I'm wrong, as the pearlescent blue looks stunning.

 

I know people who used the Feser UV and had motherboards destroyed when it leaked out. It's 10 x more electrically conductive than Feser bi-distilled water which by the way proved totally safe when it sprayed out all over my 1 day old motherboard and ram while it was switched-on.

These days I use Thermochill EC6 which has also proven to be electrically inert. I don't like using water because u need to add anti-corrosive and biocide. Easier to use something like Thermochill.

 

The ice white looks kind of... um... gross. Let's just say I wouldn't want it springing a leak.

And the pearlescent stuff (hope there's no white version of that) is really just for shows and for photos. :/

 

I wonder if that pastel version stains the loop after a while? Blocks especially...

 

Agreed
I cant see the problem people have with it, never had any gunking problems with it but then I change my coolant and hoses at least once a year when a component gets upgraded.
Never had any issues with it conducting either, accidentally spilt a load over the back of a graphics card with no problem and more recently had 2 frozenq reservoirs (v series) with ccfl's in the centre of them, have the light tube crack and leak coolant directly onto the ccfl, had it been 'normal' coolant there would no doubt have been a loud noise followed by my pc cutting out, instead of it happily running until I noticed the large puddle forming, oops

 

I'll just stick to distilled water at $.60 a gallon. The 2 degree difference is not important enough for me to spend 10x more for special fluids.

 

A trick I learned when I worked as a chemistry technician: if you want to be sure you've rinsed something out really well, give it a final rinse with the stuff you're about to fill it with.

 

I'm still waiting for blue ones because if I don't find the proper blue colour that I'm looking for might as well go with more expensive blue tubing and using simple distilled water with additives.
Come on BT, put some pressure in the companies to provide some blue coolant

Nice to see that the characteristics to look for in a coolant are price and colour. Simpler than that isn't almost impossible

 

If you want it to be blue, is there a problem with just adding some blue food dye to deionised water?

Serious question; I honestly don't know if you can get away with that or if it'll deposit out all over the fittings.

 

Fluorinert can't be used with most flexible tubing including plastics as it will permeate most of them. It pretty much demands metal tubing, and a pump which doesn't have plastics in contact with the fluid. That means that to compare it with other coolants would require a more expensive/complicated setup.

 

The pastel range and Aurora range both should be oky up to 2 weeks of being installed. we actualy had it (totally by accident) dripping on a cpu for 3 weeks with out noticing it until l we looked at the fluid an wondered why 1/2 of it was missing out the res.

Pastel Ice White will not stain any look as it doesn't contain dye how ever pastel colour will over time. Pastel is a nano fluid. Pastel all so can be used in a system for 2 years plus with out the need to be changed.

@Phil blue food dye off the shelf (super market) doesn't just contain dye it all so contains additives and sweeteners. So if you defiantly want a off the self food dye please go a head and when you've blocked your system fully and blow all your hardware you may revise you thought on that idea.

 

Does it? Someone in here built a rig with fluorinert as coolant and I don't remember him/her using any specifically metal parts...

 

There are a few problems with fluorinert in a Normal PC.

1) It expands rapidly and unless there is a vent value in you system it will crack your res and other parts.
2) fluorinert is 4x heavy than water
3) fluorinert Is not environmentally friendly.
4) you need a decent pump to move it around.

Now 3M all so make Novac which expands less and all so is not as heavy as fluorinert but it all so has specific heat range as well that it can and should be used in.

We can get both products and ive been lucky enough to test both of them as well how ever fluorinert is abut £35.00 per kilo and Novac for the good stuff is about £80 per Kilo.

 

Can the author that wrote that article explain to me why he has 3C of systematic error in temps? It is an irrefutable fact that pure water would have best temps, any additives would have made water same (if small enough) or slightly worse temps.

Distilled water has spec heat of 4.2 kj/kg/k and thermal conductance of 0.6w/mk. Ethylene glycol values are roughly half that, propylene glycol even lower. Not to mention viscosity increase with EG than again with PG causing lower flow.

Had accurate temp measurements been done, any of those fluids with 98% water with just anticorrosives, including 100% deionized water all would test within 0.2C of each other or margin of error with normal testing employing accurate measurements like using crysalfonz interface with dallas 1 wire probes, measuring intake ambient of each fan on rad, which can reliably get 0.2C repeatability.

Any fluid that has 15% EG and rest water, will be about 1C worse than distilled, again if you measure temps accurately at tdp you tested.

But then I see fluid xp+ which has propylene glycol, which will decease temps and flow quite a bit, and measuring the same as water.

If it were random error, one would expect water to test more in middle, but to have it come in last place...requires systematic error.

not trying to be harsh, but truth is the truth. I am interested in blingy fluids for taking pics in some of my builds, and getting 0.2C worse temps doesnt bother me, but when I see a review with temps that have IRREFUTABLE systematic error, then I cant believe anything in what may have been an otherwise worthwhile review.