Nanofluid tested with three different temperatures on a Q6600

i should have been more specific, sorry. I was asking skinnee, since he has already tested it once.

Don't get me wrong the nano fluid sounds interesting, it's just that all concerns need to be laid to rest with data and pics.

Quote:

---

Originally Posted by fart_plume

How about pump bearing damage?

---

Quote:

---

Originally Posted by relttem

that was my brinell hardness response.. I will have to check on that. But, that is the way I am leaning.. I will let you know.

---

Nothing evident...and the same goes for Naekuh's request on the block internals, nothing evident. There was more build-up on the Fuzion V1 inlet (slight lip on the washer where the cover meets the mid-plate), but this is versus distilled which has absolutely no build-up.

The only major difference between distilled and nanofluid was the residue and build-up throughout the loop. Every component in the loop had residue or build-up from the nanofluid, including the funnel I used for filling.

I will be using more instrumentation on the next round of tests, thats for sure.

Quote:

---

Originally Posted by skinnee

The only thing I can really point at in terms of cons to nanofluid was the tube staining and residue that was left behind in the reservoir. To me this happened at an accelerated pace with nanofluid compared to any other coolant I've used, was also much more difficult to clean than other coolant residue/staining.

However, I made relttem aware of this problem shortly after the first test and he stated they would look into any possible resolutions for the problem.

---

That was a surfactant problem, which we are working on now. The surfactant, for those that don't know, is going to keep the nanoparticles separated in the fluid. If they agglomerate, which I have mentioned earlier as a problem, you have what is called "stokes settling", and this gets worse as more agglomeration occurs. The rate of agglomeration occurring is going to be a function of the available surface area. So, basically, the surface area doubles when two particles stick together and increases as more stick..that gives more mass, which is going to cause the particles to settle faster. But, we are definitely working on that because that, to us, is the most important thing.

Quote:

---

Originally Posted by skinnee

Nothing evident...and the same goes for Naekuh's request on the block internals, nothing evident. There was more build-up on the Fuzion V1 inlet (slight lip on the washer where the cover meets the mid-plate), but this is versus distilled which has absolutely no build-up.

The only major difference between distilled and nanofluid was the residue and build-up throughout the loop. Every component in the loop had residue or build-up from the nanofluid, including the funnel I used for filling.

I will be using more instrumentation on the next round of tests, thats for sure.

---

sorry, I thought he was asking me about that..:)

Look at this way relttem we're going to ask hard questions here and not give up till we have answers and proof. If you can show us and convince this bunch of diehards, Your sure to have winner on your hands. So just hang in there.

Quote:

---

Originally Posted by fart_plume

i should have been more specific, sorry. I was asking skinnee, since he has already tested it once.

Don't get me wrong the nano fluid sounds interesting, it's just that all concerns need to be laid to rest with data and pics.

---

I totally agree..that is why we are sending this stuff out to get tested..pretty much to almost random people. The three on here that tested we recommended to us, as where the guy from Anandtech and Toms. The three on here ran similar but different tests, which gave pretty much the same results - within statistical error. HESmelaugh was our European tester and was able to run slightly different tests than the three here. And, he also got very similar results. Which, is a good thing. Again, I am all for answering your questions and trying my best to explain stuff. The last thing I want to do is sell some crap stuff.

Quote:

---

Originally Posted by relttem

I totally agree..that is why we are sending this stuff out to get tested..pretty much to almost random people. The three on here that tested we recommended to us, as where the guy from Anandtech and Toms. The three on here ran similar but different tests, which gave pretty much the same results - within statistical error. HESmelaugh was our European tester and was able to run slightly different tests than the three here. And, he also got very similar results. Which, is a good thing. Again, I am all for answering your questions and trying my best to explain stuff. The last thing I want to do is sell some crap stuff.

---

This is why I had brought up the bearing issue, most people don't even think of things like this, but due to my professional experiences a bearing failure just ruins your day. So on a high end system this would be especailly heart breaking.

Hopefully we can think of things your people haven't, and get them all covered, minimizing any possible failures.

Quote:

---

Originally Posted by relttem

the ceramic's brinell hardness will be higher than the nanoparticles. I will have to look into this. But, I think due to the use of Copper that the ceramic would have to be stronger brinell-wise, and our nanoparticles are less than copper.

---

Quote:

---

Originally Posted by relttem

that was my brinell hardness response.. I will have to check on that. But, that is the way I am leaning.. I will let you know.

---

Brinell hardness has zero to do with wear. If you can cut steel with pure water, which i'm positive water "softer" than steel, what the hell does brinell hardness have to do with anything?

Quote:

---

Originally Posted by fart_plume

This is why I had brought up the bearing issue, most people don't even think of things like this, but due to my professional experiences a bearing failure just ruins your day. So on a high end system this would be especailly heart breaking.

Hopefully we can think of things your people haven't, and get them all covered, minimizing any possible failures.

---

I honestly never thought of it because I had thought that the pumps were all sealed bearings. But, when you mentioned ceramic bearings and water being used as a lubricant I thought about the pump manufacture that is making that set-up must have taken into consideration metal particles in the water; due to tightening of fittings, micro shavings from the block that weren't cleaned out..etc etc..and, for people that just use tap water in their system, eating up the bearings. They would have had to think of a way to get past that. One easy one would be to make the bearing harder than typical metals. That is why I thought that the ceramic bearings are probably harder than the metal. But, again, I will look into that some more.

Quote:

---

Originally Posted by gillbot

Brinell hardness has zero to do with wear. If you can cut steel with pure water, which i'm positive water "softer" than steel, what the hell does brinell hardness have to do with anything?

---

who cuts steel with pure water?

Quote:

---

Originally Posted by relttem

who cuts steel with pure water?

---

Quote:

---

Within every abrasive waterjet is a pure waterjet cutter, and you can switch from abrasive waterjet cutting to pure waterjet cutting in just two minutes.

---

The harder the material, the higher the pressure. The only limitation is the pressure of the machine. Regardless of what you are cutting, i'm sure MOST of the materials are "softer" than water, even if it's only plastic. Brinell hardness is completely irrelevant.

Quote:

---

Originally Posted by relttem

I honestly never thought of it because I had thought that the pumps were all sealed bearings. But, when you mentioned ceramic bearings and water being used as a lubricant I thought about the pump manufacture that is making that set-up must have taken into consideration metal particles in the water; due to tightening of fittings, micro shavings from the block that weren't cleaned out..etc etc..and, for people that just use tap water in their system, eating up the bearings. They would have had to think of a way to get past that. One easy one would be to make the bearing harder than typical metals. That is why I thought that the ceramic bearings are probably harder than the metal. But, again, I will look into that some more.

---

Typically water pumps that use a ceramic bearing are the type that self polish,(because the impeller shaft has a matching bearing surface). Introducing a forien object in to this surface can scar it. depending on the clearances of the bearings and the shape of the object, the shape affects the scarring as much as hardness does. ie a soft steel shaving in an automatic transmission can do damage to the hardened bearing surfaces.

Quote:

---

Originally Posted by relttem

who cuts steel with pure water?

---

Well steel shim stock can be cut with just water, but as you said why when there are better methods..............

Quote:

---

Originally Posted by gillbot

The harder the material, the higher the pressure. The only limitation is the pressure of the machine. Regardless of what you are cutting, i'm sure MOST of the materials are "softer" than water, even if it's only plastic. Brinell hardness is completely irrelevant.

---

you do have a limit to what pressure you can achieve. Now I am curious as to how much pressure you would need to cut thru a .25" steel plate. I will calculate that out tomorrow. Believe it or not, it will be a function of density..how about that?

Why is this even an issue, unless you're planning to sell the fluid outside computer cooling. The turbulence of the fluid through the pump and tubes is plenty enough to keep the particles suspended indefinitely.
If computer off for 6 months, no problem, them selttling will go away as soon as you introduce turbulence with the startup of the pump.

David

Quote:

---

Originally Posted by relttem

That was a surfactant problem, which we are working on now. The surfactant, for those that don't know, is going to keep the nanoparticles separated in the fluid. If they agglomerate, which I have mentioned earlier as a problem, you have what is called "stokes settling", and this gets worse as more agglomeration occurs. The rate of agglomeration occurring is going to be a function of the available surface area. So, basically, the surface area doubles when two particles stick together and increases as more stick..that gives more mass, which is going to cause the particles to settle faster. But, we are definitely working on that because that, to us, is the most important thing.

---

First of all: Sorry I haven't checked in here lately. I hadn't noticed that this thread had become active again.

While I can't answer the technical side of most of the questions asked (I don't have any insider-knowledge on the Nanofluid) I can announce that I'm planning on doing a long-term test with the Nanofluid. I will use it in my productive system which runs about 18 hours a day. Maybe that will help answer some questions about long-term effects.

Quote:

---

Originally Posted by Eternalightwith

Why is this even an issue, unless you're planning to sell the fluid outside computer cooling. The turbulence of the fluid through the pump and tubes is plenty enough to keep the particles suspended indefinitely.
If computer off for 6 months, no problem, them selttling will go away as soon as you introduce turbulence with the startup of the pump.

David

---

We have noticed this being a problem only with certain pump set-ups. After Skinnee told me about it, I checked my WC system and had some residue in it. Yes, the flow thru the block and tubes does keep it mixed well, but when it is in the reservoir - not so much. My pump intake is on the side of the reservoir, therefore I have a dead zone beneath the intake port. We have reservoirs that have the intake at the bottom that will not see this happen, because there is no dead zone. If that makes sense.
I believe HESmelaugh's set-up has a bottom intake. If you look at the pictures of his set-up you can see that. I am not sure what kind Skinnee had. Also, I have had the nanofluid running in my WC system since June, 2008.

Quote:

---

Originally Posted by gillbot

Brinell hardness has zero to do with wear. If you can cut steel with pure water, which i'm positive water "softer" than steel, what the hell does brinell hardness have to do with anything?

---

Because Brinell scale measurement measures the indentation ability of a material. The virtual hardness of water is massively increased in cutting machines in jets.

I'd rather use this than some white tubing as there's always certain advantages when you can actually see the coolant.

Nice testing, but I'd also like to see some passive rad tests:up:

Quote:

---

Originally Posted by relttem

I really don't have an answer for the 3M question. I don't think 3M really does any sort of cooling stuff. DowTherm does, but their niche is not electronic cooling. Also, this nanofluid stuff is new technology, maybe they just haven't gotten around to it yet. We were lucky that we had the Air Force to sponsor us.
I appreciate all the questions and comments - it keeps us honest.

---

3M makes FlourInert and HydroFlouroEther. Both are completely non-conductive. And used to directly cool electronic components and motors.

Both do not cool as well as water, but are a necessity in some applications. But then again in those applications you can use a refrigeration system and a heat exchanger.

Speaking of which, my AC guy should be here soon to fix my refrigerator for just that application.

http://en.wikipedia.org/wiki/Fluorinert

3M makes almost everything. :rofl:

Quote:

---

Originally Posted by Lagaaja

I'd rather use this than some white tubing as there's always certain advantages when you can actually see the coolant.

Nice testing, but I'd also like to see some passive rad tests:up:

---

We've been trying to get some passive radiator testing done. But, we need a radiator that has fins that run up and down as opposed to left and right. When we find one of those we will give it a go.

Thanks for the info on 3M - I had no idea.

Quote:

---

Originally Posted by relttem

We've been trying to get some passive radiator testing done. But, we need a radiator that has fins that run up and down as opposed to left and right. When we find one of those we will give it a go.

---

Now you ask for one... just threw away one about 4 days ago that would have been perfect for what you wanted (already had it modded for pc use too).

Quote:

---

Originally Posted by relttem

We've been trying to get some passive radiator testing done. But, we need a radiator that has fins that run up and down as opposed to left and right. When we find one of those we will give it a go.

---

Just turn the rad around....:rofl: Yeah I know I am a smartass!

Quote:

---

Originally Posted by Sadasius

Just turn the rad around....:rofl: Yeah I know I am a smartass!

---

actually, well...I will let you figure that one out..:)

I have an update on the subject of Nanofluid. I got a new sample (new formula) of the fluid for testing.
I decided to do a very down-to-earth practical test of the fluid this time. I used it in my business PC, equipped for this purpose with a GTX 285, so that it produces a bit more heat.

Here are the components making up the loop:
- XSPC X2O 450 pump
- Watercool 360 radiator
- D-Tek Fuzion V2 (on a Q6600 @ 2.7 GHz, 1.3 V)
- EK FC-285 GTX (on a PNY GTX 285 @ stock)
- 1/2" tubing
- Some thermal probes

As you can see, the loop has a relatively underpowered pump (nice and quiet, though =) ) and a radiator that is actually a bit too small for the CPU and such a powerful GPU.

For the test, I first ran the loop with distilled water. I ran Prime95 (small FFT) and Furmark for 45 Minutes and measured the temperatures. Then, I changed to Nanofluid in the loop, did the same test and measured temperatures again. Here are the results, shown in temperature-differences between component and fluid temperatures.

http://www.abload.de/img/newnf1a8i3.jpg

As you can see, the Nanofluid does make a measurable difference, here. I must add that the error margin for this test is expected to be higher than for the tests I run on my test-bench, since I don't have the same, precise measuring tools in my working PC. The results still show a positive trend, though.

I am now running my PC with the Nanofluid and intend to keep it like this for at least a few months. Then, I'll take the loop apart to see if there are any long-term impacts from using the Nanofluid.