Coolant/Fluid Roundup – Thermal Performance

A topic that seems to resurface every couple of weeks around the forums is what coolant/fluid is best; this usually ends up turning into a heated debate of opinions with no data to douse the flames. There is an ever-growing variety of coolants that are available in a rainbow of colors, which is great for those of you trying to achieve a certain color scheme where colored tubing just will not cut it or you have nightmares when your reservoir contains a clear fluid that just does not match your color scheme.

A variety of fluids on the market have been marketed as anti-corrosive, non-conductive, earth friendly and the list goes on and on. The anti-corrosive feature potentially holds weight for the times when you just cannot avoid a little aluminum in your loop, but the “no more aluminum” crusade eliminated nearly every aluminum product out there. Copper, brass and even silver make up our cooling products now, you have to search out aluminum on purpose to find a product that has it in the flow path. Non-conductive, well, all fluids will become conductive over time in your loop… so there goes that marketing angle.

So what are we left with you ask, color and thermal performance are the only things left in my mind. Those of you who frequent forums have all viewed the threads where colored fluids have broken down and clogged up micro-pins or impingement plates. Additives in our fluids will break down, this should be of no surprise, the piece not discussed on those gunk/clogging threads is that the coolant breaking down contains more than just water and color dye. I am of the opinion (opinion because I have no hard data on the matter) that the additives besides the dye are the real culprit.

Last but not least is performance, and that is exactly what we were on a mission to find with this round of testing and review. Let us not waste any more time with my introduction and get to the matter at hand…

Included in the roundup is distilled water with PT_nuke, Feser One, Feser Ultra Pure, Fluid XP Nano-Fluid, Ice Dragon Nanofluid Formula C2, Ice Dragon Nanofluid retail formula, Minnesota Tap water, Primochill PC Ice, Primochill PC Pure, Swiftech Hydrx and Thermochill EC6. We tried to cover as many of the latest trends, popular fluids, and classics as possible; we even have tap water fresh from my city water supply. Nanofluids are the latest advancement looking to knock good old distilled water off the hill; we have three, the early prototype formula from Ice Dragon (Formula C2), Ice Dragon retail, and the most colorful selection in our roundup Fluid XP Nano-fluid.

Providing the primary backdrop for our coolants is the box from our X58-UD7, Gigabyte sent over the UD7 for the several system level tests we have planned and the coolant roundup is the first. We will cover more on the UD7 as we progress with the system tests we have planned, but with a little preview of our test system, time to move beyond the intro and dive into our test specification and methodology.

Intrigued? Head over to Skinnee Labs for the full article…

Loading new replies...

Avatar of Owenator
Owenator

Senior Internet Fart

1,486 messages 13 likes

Very nice article!

I'll be using distilled water with biocide from now on.

Reply Like

Avatar of I.M.O.G.
I.M.O.G.

Glorious Leader

25,037 messages 3 likes

Yea, Skinnee is the authority on good cooling testing these days, and I'm glad to see some solid numbers on this... I wouldn't have guessed the variation between fluids would be so small.

I'm still having a hard time believing Fluid XP is that much different, but I can't see how the results could be wrong given the methodology. Its just so much worse compared to the others, its hard for me to comprehend how you could make a coolant/fluid that much different in light of how consistent all the other products are with eachother.

Reply Like

Avatar of David
David

Forums Super Moderator

15,810 messages 10 likes

Interesting, all the GPU stuff is within ca. 2 standard deviations (with the exception of Fluid XP), suggests that there really isn't much difference between any of these solutions....

Do you have any idea what your error is? How accurate is your thermocouple?

Reply Like

Avatar of m0r7if3r
m0r7if3r

Member, Water Cooling Sticky Reading Enforcement O

8,986 messages 0 likes

Interesting, all the GPU stuff is within ca. 2 standard deviations (with the exception of Fluid XP), suggests that there really isn't much difference between any of these solutions....

Do you have any idea what your error is? How accurate is your thermocouple?

I think vapor said .125C...but I could be wrong...i'll let skinnee give you a definitive answer

EDIT: well that was quick, found vapor's post (forgot he wasn't under vapor) http://overclockers.com/forums/showpost.php?p=6618292&postcount=18

Reply Like

click to expand...
Avatar of RGE
RGE

Senior Member

1,029 messages 0 likes

Fluid XP nanofluid (ion scavenging nanofluid, ie not an attempt at thermal enhancement) is the only one that is at least 50% ? propylene glycol, based on their web page description of designed for fuel cells and chemical properties description including specific heat capacity, boiling, freezing, etc, ie it is likely this stuff.

All the other liquids from their description I believe are 85% water or higher, probably main reason why Fluid XP nanofluid is not with rest of the pack. Guess you pay a big thermal performance price for the extra weeks of slowing electrical conductivity.

Reply Like

b
bcsizemo

Registered

19 messages 0 likes

Scratching head....

I'm not sure why this is really a big surprise...

Water has one of the best thermal capacity vs. thermal resistance of any "cheap" liquid. There are only 3 reasons you really even add anti-freeze to your radiator in the first place: boil over, freezing, and anti-corrosion/pump lube. If you lived in a place where it was always 50F you could probably run straight water and a little water wetter.

Now much like the liquid nitrogen guys I'm sure 3M or someone has a hydrocarbon based liquid that has some better properties. And I'm sure it's probably in that magical range of $1k per liter or something. That and it'll probably only get you %10 better theoretical specs...

I guess I've just been out of water cooling to long. I didn't even know companies made things like this. I thought most of the time it was just water or water/water wetter (or your additive of choice).

Reply Like

click to expand...
Avatar of skinnee
skinnee

WB, pump & rad molester

163 messages 0 likes

Thanks! :thup:

With the sensors deployed on the bench, +/- 0.125C is as good as we can get for the system.

Reply Like

Q
QuietIce

Disabled

15,734 messages 0 likes
Avatar of macklin01
macklin01

Computational Oncologist / Biomathematician / Mode

5,663 messages 0 likes

Excellent

This is computer journalism at its best: evaluating whether there's any appreciable differences between these (often expensive products) and just simple distilled water + biocide. This is the type of reporting that causes us all to step back and re-evaluate our buying decisions, and it provides a great deal of value to the community.

I can't stress enough how important I feel it is to take a critical look at our own practices, to see what is worthwhile and what is not. And this is exactly what you have done.

Bravo! More of this!

Reply Like

Avatar of nd4spdbh2
nd4spdbh2

Member

9,212 messages 1 likes

i would really like to see some tests under the same conditions using distilled water and water wetter... the stuff used in car radiators that reduces surface tension... i always use the stuff in my motorcycles and it definitely drops temps quite a bit... but we are dealing with much higher energy transfers / flow rates and temps... so it might not have nearly as much of an effect on something like a comp.

Reply Like