Coolant Testing By Skinnee

http://skinneelabs.com/coolants.html

Haven't read it yet, interested to see the results though.

EDIT: just finished reading, I think this image speaks for itself


EDIT2:Chart changed, as promised.

damn, I just got the thing published and haven't even had the chance to add the article over here.

Hah, you're in my startup tabs, so i check pretty frequently lol

lol, no problem with that.

Distilled 1 other coolant 0.

i dont get it? isnt the one with the ZiO in it Hotter?
i was trying to follow all this insanity before knowing nothing about it, and was wondering some things. you know Zinc Oxide is a powerfull algicide and moss preventor and all. and the claim was it was neutral boyant, but it sinks in my water (as used as a moss preventer)

and then i was wondering about zinc sulfides, like zinc monohydrate and all which mixes right into water (not particals floating in it), and things like copper sulfides which do the same, but the PH on those is not normal, so i wonder what that would do to an uncoated cooper or aluminum surface. both of those would also prevent algae formations.
plus if you look at the GOOP on the block for that freezer stuff, it looks more like burnt protiens and carbohydrates than a dye, which could indicate algae, any time the algicide is no longer fuctional in any of them, be it a silver strip or metal oxides, algae could begin to grow, and would string out, and ketch in the block and die there. algae lives at certain temperatures.

i also wonder why there isnt some UV leds in some of this stuff to do a UV killing of any live stuff slowly and regularly to kill any live stuff to keep it from growing.
.

Grosjambon said:

This Testing Killed Something
Distilled 1 other coolant 0.

Click to expand...

Being that the closest one is only off by 0.1C, I wouldn't necessarily say it kills it. The thing that kills it is the fact that these other fluids tend to break down in the loop and clog pumps and blocks.

Psycogeec said:

i dont get it? isnt the one with the ZiO in it Hotter?
i was trying to follow all this insanity before knowing nothing about it, and was wondering some things. you know Zinc Oxide is a powerfull algicide and moss preventor and all. and the claim was it was nuteral boyant, but it sinks in my water (as used as a moss preventor)
and then i was wondering about zinc sulfides, like zink monohydrate and all which mixes right into water, and things like copper sulfides which do the same, but the PH on those is not normal, so i wonder what that would do to an uncoated cooper or aluminum surface.

Click to expand...

@ deadlysyn IT stil win Even by 0.1 Degree

Both Nanofluid C2 and Nanofluid R have Zinc Oxide as the nano-particles suspended in water.

As for all the other zinc's you mention, I have no idea.

skinnee said:

As for all the other zinc's you mention, I have no idea.

Click to expand...

neither do I, i just wonder if your trying to get "salts" into a fluid to trasfer heat faster, or "metals" in safe forms, all of which is done in other areas of (what do we call it) "fluid dynamics" or some such cute name, it seems like they would choose soluble items?
but just like i am not an EE , I am even further from being a chemist.

Grosjambon said:

@ deadlysyn IT stil win Even by 0.1 Degree
Off Topic ?

Click to expand...

While 0.1 degree is still a win, it's nothing huge, but we still know the truth about what these other fluids do to our loops. It really is a shame that it would take so long to test the long term effects. I'm still sticking by my decision to use distilled and True Silver barbs. I don't see that changing anytime soon, unless I see algae starting to build up, then I will be doing a complete tear down, and ordering up a bottle of PT Nuke.

Edit: I just wanted to add that over the long term, when these other fluids start breaking down and clogging up the blocks, the temperature margin can get quite large from what I have read. It seems that some have had their temps go up by close to 15C just from the break down alone.

Yes, the more additives in the fluid the more there will be to break down and gunk up your loop which will in turn cause lower block performance and potentially flow rates. How long is the time span in order to reach the point of break down... I don't know if we'll ever get an answer on that just due to the resources required to find out.

@Psycogeec - well said!

skinnee said:

Yes, the more additives in the fluid the more there will be to break down and gunk up your loop which will in turn cause lower block performance and potentially flow rates. How long is the time span in order to reach the point of break down... I don't know if we'll ever get an answer on that just due to the resources required to find out.

@Psycogeec - well said!

Click to expand...

The more I have been thinking about this, with all the talk of different fluids, I think the amount of time may vary from one system to another. Say I build 2 identical rigs, same CPU, same stepping, same board, same RAM, etc. Is there actually an explanation besides the old one about how no 2 systems are the same when it comes to why one clocks higher than the other, or why one runs hotter? I think the point that I am trying to get at, is that there is likely to really be no for sure amount of time before breakdown occurs. There are too many variables. Maybe it has something to do with heat. Maybe it's light, or maybe just suspended particles getting caught on the pins and grooves in the blocks. If we specifically knew what the cause was, maybe it would be easier to test for an estimate of time.

I think you have it spot on, the number of variables to control in trying to capture the time for a fluid to breakdown is almost impossible to capture, let alone control. However, we do know it happens as we've seen the photos.

All the thinking I did on coolants preparing for testing is what lead me to write the intro the way I did and come to the conclusion that the best use of time and resources was a look at thermal performance alone. But I had to talk about all the other factors... which I think I accomplished.

Great write up. Glad you got it done and pretty much verified that all excluding one cool very close to each other. But stating which is the easiest and less hassle.

Well done!

How is the wattage data pulled? Can you post anything as to the veracity of the higher wattage meaning more workload statement? Also, where did all this data come from, i know you said skinnee sent it to you, but i saw nothing of the sort in the article, was this just data he collected and didn't post? If this is the case, I believe a bit more testing is in order to see if the increase in wattage actually has any significant effect on computing performance, even if it is only on a level that would be applicable for benchmarking. I don't mean to shoot holes in your analysis, there was just a little in there that wasn't satisfactorily explained.

Any verification on the increased chip efficiency, theoretical or applied? I'm not discounting it, just interested.

also, i'm interested as to how the watts dissipated came about. I think that's where the primary problem lies, the fluid interactions, the fluids that transfer heat across the fluid worse will have a higher delta for the rad in/out where the fluids that reach equilibrium within the fluid will have a lower delta...it's not representative of the thermal output of the chip to me, but rather the speed at which the body of fluid within your loop can reach equlibrium...maybe i'm thinking about it wrong though...

relttem, it looks like you're just making up numbers to make your fluid look better. Saying Nanofluid "had an average increase in load of 16% to that of the Tap Water test" is utter BS. There was no such change in load. Look at the Air Delta figures if you want to see any variation in load/measurement--they range between 7.59C and 7.95C with Nanofluid right in the middle.

An error of +-.125c in the temperature measurement easily explains the 16% difference in heatload you're making up (and which isn't seen in the air deltas). Please don't manipulate the numbers we post to skinneelabs for your own purposes.

The data is pretty clear, IMO--water leads the pack for coolant performance and the various premixes are basically just as good (since they are basically water with some additives) with an exception in FluidXP Nano-Fluid (which isn't just colored water).

-Vapor from skinneelabs and XS

relttem said:

I don't follow the question about 'cjip efficiency'..what do you mean?

Click to expand...

I mean this

relttem said:

That lets us calculate the load applied to the system using Q=mdot*Cp*dT. Interestingly enough, there was a variance in load. I would never have thought about this if I was running the same benchmarks etc. I actually talked to Ed Calloway at AMD about this, and he said it was due to the chips dynamics trying to control what is going to each core.

Click to expand...

you claim that since your fluid has a higher dT across the rad, it is making the chip run more efficiently, or did i misunderstand.

relttem said:

naw, I am not claiming that by any means. I didn't know it was possible to have a CPU put out different amts of heat when doing the same operation. I would have thought that it would always be the same. But, evidently it is not, as shown by Skinnee's data and by what what ed said.

Click to expand...

Oh...ok...can you verify this? Also, in that, I lost the meaning of your first graph :S