Modding Watercooling Fluid Shootout

Another cracking article Brett. I would like to see a proper test of 'non-conductivity'. I've had small leaks of distilled onto some of my components and its never done any harm to them, so until proved otherwise i would doubt the benefit of any of these coolants over water. Can you not set up a test of a 'catastrophic leak' and have it dump all the coolant onto some old hardware?

Alos why did you use water wetter as the control additive? I was under the impression it had been rejected by the watercooling community as it deposited goo into your loop. CandyKid on Wizdforums had a very informative post on the subject.

 

see the post right after yours.

but also, if you know what is in the dyes, simple chemistry will be able to tell you what will neutralize that.

 

Not sure what you're pointing me to here? Are you saying that NCF are pointless? That may be the case but I'd rather not take the risk. It may seem a naive attitude to you but having heard many tales of Feser One spillage causing zero collateral damage, I feel slightly more comfortable with it whizzing round my loops. Particularly after hearing horror stories of distilled/dionised water spillages (when mixed with the necessary corosion inhibitors and UV dyes) having more adverse effects.

When it comes to cost, as Brett mentioned, it's differs little to distilled/deionised water, when you take into account the additives. Also, with it performing almost as well as distilled water, why take the risk?

As I said, I have no knowledge of chemical properties/reactions/effects so regardless of how "simple" the problem may seem to yourself, I might as well be trying to split atoms.

 

Hey Da_Rude,

Thanks for the love. Now, as to your questions.
1) I'll try it to see if I have anything friable that's not already fried. To say that Bit-tech puts a strain on hardware is, well, a complete understatement. I have old junker boards, but for the most part they're junk because we use and abuse. But for the most part, you need to consider it like this - these fluids are designed to prevent a leak from blowing up your hardware - NOT to be a submersible chemical. It is not totally inert - it just doesn't conduct 12v or below very well when you aren't talking a power-supply worth of amps/wattage. Even a substantial puddle isn't going to kill things, but if you douse the board in it then you may not be as happy a camper.

2) I used water wetter as a readily available and effective compound found in many watercooling loops. Yes, it leaves goo over long periods of time - but the 2hrs it was in my loop wasn't going to put down too much sediment. I don't advise using the stuff, tbh - but that's because when there's as cheap and good of an alternative in Feser, I can't think why anybody SHOULD use it.

Heh. I've had years of proper University chemistry couses at one of the leading US uni's for chemical research - and I don't think that I'd wanna try that any more than you. Besides, most of these compounds are copyright/patent blah blah blah - so you'd be hard pressed to know just what's in them unless you have an NMR machine at your disposal. Maybe Bluephoenix does, though - so hey, if he can do it, he should post the results for the rest of the class, too.

 

All I can say is:

Distilled Water = Free and no environmental side effects (not that any of these other products have any but some do)

If you are careful and pressure test your cooling system before using it, you won't have any problems and if you do they are usually covered by warranty. Now i guess it's nice to have the added security of knowing that these coolants won't damage your system in the event of a failure, but for me I just don't see the cost benefit.

 

Hey Scootiep,

It's an interesting point, but I just wanted to bring a point to readers -
Warranties, however lenient, will NOT cover liquid cooling leaks. And if they do, they are never more than for the value of your cooling purchase. Further, it is nigh on impossible to determine if a leak in a cooling setup is user-error or product error. This makes it even less likely that you'd ever see a dime.

All this translates to - if your loop leaks, for whatever reason, you're likely screwed. Some people will say that "if you build your loop right, it'll never leak." I, on the other hand, look at a non-conductive fluid as insurance. If I can spend 8 quid to make sure I don't lose a 300 quid graphics card, while looking cool to boot, I'm all for it.

 

Ah, so another "expert" with chemicals suggests it's not quite as "simple" as Bluephoenix suggests. I had my suspicions but like I say, I'm a chemics n00b

Whilst I agree with you in principal Scootiep, I don't necessarily agree in practice.

Distilled water is (or rather, can be) free, but without an anti-fungal solution and possibly corrision inhibitor (depending on if you're using mixed metals) it will need regularly changing and lead to cloudy tubing. Plus as gimmicky as it may seem, you gotta love the colours . When you take into account those additives (that lets face it, most people want or need) Feser One is the better option for the majority and doesn't cost much different.

Even the most carefully pressure tested loop can develop leaks after time. Granted a thorough test will hopefully ensure no major leaks but it doesn't stop a small one developing days/weeks/months after the installation. You say: "you won't have any problems and if you do they are usually covered by warranty"; that may apply to the components themselves (be it radiator, block etc) but it certainly won't apply to any components that may be damaged in the event of a leak (mobo, cpu, gpu etc).

Your final point regarding cost benefits, well, I think my last two paragraphs sums up what I think about that.

EDIT: Bah, Brett beat me to it.

Any thoughts on the UV "reactiveness" of the Blue Feser you used btw Brett (see earlier posts)?

 

Heh..I'm far from an expert - but I've had a pretty good education, as both specofdust and bindi can attest.

As for the UV reactiveness - the guys are right that the UV blue stuff does not glow like a proverbial christmas tree, but that's part of it being blue. UV reactive stuff is "less reactive" (not really, but visibly) the closer it gets to the UV spectrum - so blues and purples will always be faint compared to reds and greens. It's an unfortunate reality.

 

LOL, don't worry Brett, I don't need to see your credentials, I'll take your word for it

Um "the guys" were.. me.

*Looks up "UV Spectrum" on google images and realises it's the same as a rainbow (with an extra UV and IR bit added to each end)*

Ah, I see

So theoretically, red would be the most reactive, followed by orange, yellow and then green. Jesus they must be insanely bright!

 

It's a bit more logarithmic than that, but...yeah. If you look at an orange or red piece of sleeving, for example, vs. blue. Best to do this with a cold cathode in a lit office. You can see the red fluorescence under a UV light in pretty bright office light, whereas blue would be much less visible. And don't forget, our eyes are most sensitive to green light, which means that even if it's not giving off as much fluorescence, it is *perceived* to be almost as bright as the reds.

NOTE (for the flamers ready to pounce): This is simplified explanation - properly, we're talking about the luminescence of the spectrum of available fluorescence due to near-UV spectrum Woods blacklighting. I am not going to be this pedantic to muddle the issue. I just want to stop the next guy saying "wait, this isn't the proper UV spectrum, j00 don't no nuffin!"

 

Cheers for the explanation Brett .

I wouldn't worry about them Brett, they'll be too busy pissing themselves laughing at me to question the validity of your statements

 

Fluid XP is NASTY stuff. I tried the red stuff a while back, and within a few months, the entire liquid began to destabilize and decompose. Solids start to precipitate and clog everything up, and the entire loop becomes cloudy, mucky, and stained. This stuff doesn't just stain the walls of the tubing as other dyes do, it cakes stuff on there.

Stay away from Fluid XP. The only good thing about using it is that it will convince you to just stick with distilled water and minor additives. I hope Fluid XP drowns in their dishonest claims and lousy products all the way to bankrupcy. They're almost as chavy as Silenx.

 

I would like to know about the leaks in a dusty pc. I unfortunately have a couple long hair large dogs, coupled with hard flooring, and a big back yard, creates a dust bowl in my rig. I tried using filters, but they clog up so much that it just wasn't worth the effort. And now that I've upgraded to a better, (hotter), cpu, I am leaning more to going water cooled. As my rig is an old Antec SoHo server case, (go-kart frame I believe was the review for it), is so heavy, I fear coupled with water lines, I might neglect dusting even more than I do now. So when these NCF's hit a dirty board, how much does that affect the NCf?

 

Most likely these fluids are from 3M, they make a variety of dielectric fluids for heat transfer applications. That they are not as thermally conductive as water is a given, that they will not fry your system is their primary value. Even RODI water will conduct electricity, once it contacts a board or connector it will pick up contaminates and ionize, there is no such thing as safe water.

 

A very informative article! At least for myself and all other watercooling noobs, I now have a much better idea of what to use than before I read the article.
The experts can argue all day over the UV lumious spectrum, flux capasitors and lithium crystals in your loop, but then I don't think that was the basis for the article anyways...

 

Your test only lacked one important factor: Will it wash out of you favorite shirt?

 

And once it's in your loop it's going to pick up ions or trace metals from blocks, and overtime it will BECOME conductive. Same with dionized water - there may not any ions in it when you get it, but the water will strip them out of the loop.

 

Does the ambient temperature change how affective any of the NCF are?
It will obviously change how hot they all are, but maybe some are better at transferring heat when they are hotter?
Sorry, it's been to long since I did any physics or the like... But I have been considering water cooling and live in Darwin... Average daily temps of 30+C this time of year.

 

Umm, how does the Feser One hold up against PC ICE? Here in the states, they cost about the same (probaby because the dollar is worthless now) and I'm not sure which is better for the rig I'm planning to make.

 

So we should all know now that pure H20 isn't conductive. Only conductive elements in the water are. So by using distilled water we introduce pure water into the loops (ideally). The water may then, over time, become more conductive due to the conductive impurities in the loops (contact with metal, etc). Logically, the same process should apply to any other liquid that is "non conductive" in that it may become conductive if contaminated within the loops. Therefore, both water and the mentioned products should be realatively equal in conductivity (hypothetically), and neither should be "safer". If you happen to spill either one on your board then the determining factor should be the amount of conducting contaminants on your board. Not the liquid.

By the way, there is an easy test for this theory (even though I will not be trying it).
1. Take a portion of water and test it's conductivity.
2. Add a portion of conductive material to water (table salt, sugar, etc)
3. Test the water's conductivity
4. Repeat 1-3 for product

I would actually be interested to see numerical results of a test like this, if done properly using identical portions and testing procedures (ie. same distance between test leads, etc).

This does not take into account algae, appearance, or anything other than the conductivity. Although, theoretically again, if the water is pure then algae should not be a threat.

Enjoy.