nanofluid!

[littleowl]

ok so a guy got on another forum I get on a while back and asked people if they thought of trying nanofluid and a member jumped all over him so to speak.
I PMed the guy and said I would like to know more about this stuff I am willing to see how it works for my systems. I crunch so my systems are always 100% load.

I have a cheap mobo ECS G31T-M v-1.0
With a E6600 running at the time stock speed.
The waterblock on this chip is my OLD maze4 that I made a hold down for, This system was crunching right along with core temps 32c, 32c and room temp at 23.9c again 100%load

So I got all my loop drained and put the nanofluid in and decided at the last min that the stuff looked kind of heavy so I added a little maybe 8-10 OZ. of distilled water. Boot the system back up and room temp was 25c and still stock speed with core temps the same 32c, 32c

I hurt my thumb and have not had time to mess with it at all so it has been chugging right a long in my barn. about 3 weeks go by. I decide to check on it. Well room temp was down some but didn't check it. I decided to up the FSB a little. I took it up making the chip run 2.777. One week later my I go look at it. Room temps are 28.9c my cores are 36c 36c. Still running 100%

I cant say much about if this stuff is great or not yet but I can say it helps keep my temps down when I took up the FSB. I expected it to go up 1 or 2 degrees. That is what normaly happens but it didn't.

I have a couple pics I added to this in the WCG section here (http://www.xtremesystems.org/forums/showthread.php?t=202442).

[evil-98]

what is nanofluid? it white?

[BlueAqua]

What's nanofluid? Please fill us in.

[Eddie3dfx]

There was a watercooling part on ebay a while back with some berium liquid or gas inside of it.
Is that what you are talking about?

[littleowl]

I don't know a lot guys you maybe able to learn more about it by using google.
Investigation of Increased Thermal Conductivity of Nanofluids

Fluids containing suspended solid metallic or nonmetallic particles (nanofluids) display enhanced thermal conductivities. The aim of this study is to identify and understand the mechanisms behind this increase and to develop a suitable model to predict the thermal conductivity for a variety of nanofluid systems.

There are microscopic pieces of copper in what I am running. The guys that sent it to me were testing and mixing the stuff for military applications. one guy is interested in water cooling and decided to dump it in his system. He said he is seeing some of the same things I seen so far. didn't change anything at all untill I started OC then I seen my temps didn't move. I am sure it needs pushed more but I have some troubles doing that at the moment. (darn bum thumb)

[BlueAqua]

From MIT:

Nanofluids are engineered colloidal suspensions of nanoparticles (1-100 nm) in a base fluid. Common base fluids include water and organic liquids. Nanoparticles are typically made of chemically stable metals, metal oxides or carbon in various forms. The size of the nanoparticles imparts some unique characteristics to these fluids, including greatly enhanced energy, momentum and mass transfer, as well as reduced tendency for sedimentation and erosion of the containing surfaces. Nanofluids are being investigated for numerous applications, including cooling, manufacturing, chemical and pharmaceutical processes, medical treatments, cosmetics, etc.

It's really hard to beat water in specific heat capacity, and that's important for water cooling. I'd sure love to see something that could beat it though.

[littleowl]

yeah that is a good explanation :)

I didn't learn much before I jumped into it I just wanted to make sure my block and pump wouldn't get ruined.

[SNiiPE_DoGG]

yeah that is a good explanation :)

I didn't learn much before I jumped into it I just wanted to make sure my block and pump wouldn't get ruined.

So where can I get this stuff, I'll test it :up: maybe send some over to martin too

[DarthBeavis]

fluid so has it in beta now. If someone wants to test It I can probably help .

[littleowl]

I would like to get martin to test it!!!!!! :) I will talk to the guy and see how much he has and all that.

[Bojamijams]

So metal particles inside a liquid.. hope you don't get a leak cause oh boy.. conductivity :)

[littleowl]

it is no more conductive then water.
The metal is so small that there is almost no way for them to make an arc.

[rosco]

Such white color indicates that it is, @ 99% sure, alumina nanoparticles (aluminium oxyde), probably the most classical and studied nanofluid. It doesn't beat water in terms of specific heat capacity, but on thermal conductivity (up to 30-35 % filled @ 5% v/v) which leads to a higher convective coefficient. Several thermophysical aspects are involved into the fluid efficiency, not only the specific heat capacity as said in a previous post.

[CyberDruid]

Is it UV reactive?:D

[BreeSpree]

Basically sounds like the same idea, of some people sprinkling silver into their res. Doesn't sound like such a bad idea when it comes to thermal capacity, however I do fear for the stability of a pump. I know the particles are small, but it still isn't like pure water, which is what the pumps are designed for. I wonder if the particles would speed up corrosion process too. Doesn't sound like a bad idea, just maybe if all the world's coolant companies that have scientists available to test and sell various coolants for the purpose of computer liquid cooling would have discovered and implemented it if it was such a great idea. Then again, all they sell is basically distilled water with antifreeze.

[Bojamijams]

it is no more conductive then water.
The metal is so small that there is almost no way for them to make an arc.

A lot more conductive then distilled water though.

[cegras]

Why is specific heat capacity so important?

I thought the purpose of water cooler was to wick away the heat faster than air cooling and dump it faster as well. For that, obviously a liquid with maximum conductivity would be good. And if the liquid gets hot, even better right? That means at the radiator, there's a larger T gradient for it to dump heat faster.

Of course, this complicates how many heat sources you put into the loop ..

[BlueAqua]

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

Specific heat really has to do with how much heat can a material hold. In water's case, it can hold over 4 J mol−1 K−1.

There are a lot of factors, more the specific heat capacity, but I think it's a fairly major one.

[littleowl]

A lot more conductive then distilled water though.

nope

[cegras]

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

Specific heat really has to do with how much heat can a material hold. In water's case, it can hold over 4 J mol−1 K−1.

There are a lot of factors, more the specific heat capacity, but I think it's a fairly major one.

Well yeah, I understand. But, say, given a liquid metal or nanofluid as the working fluid, I would take that over water as it would certainly conduct way way better with a not as proportional drop in specific heat capacity.

[bluphysted]

In summary...

Distilled water makes a lot more sense.


Flippin' nerds.
;p

[littleowl]

@ bluphysted :rofl: no one has good good testing to say distilled water makes better sense yet. :)

[bluphysted]

True... true.

I may be jumping the gun here.

But..

I can get 4 litres of distilled H20 for $0.99.. and.. well... I don't even know what nanofluid is, let alone where to buy it.

[littleowl]

it is not in the market for us to buy yet.

[bluphysted]

In summary...

Distilled water makes a lot more sense.


Flippin' nerds.
;p



Ok then..

like I said.

[Pedalmonkey]

could be great for a lot of people who put a lot of stress on their cooling loops for crunching farms. Lets get it to Martin or someone else capable and willing to test the stuff. could be the next jump up we've got on those air cooling freaks :P

[xtclocker]

Interesting testing littleowl, and thanks for the testing with these new whity stuffs. Hope you can still keep it in the loop to see a longterm stability and such.

[xtclocker]

sorta looks an feels like mercury, i only see trouble with pump seals,bearings for nano fluid lubricity is poor.

to tweak this need high pressure hermetic loop

Good thought, and nice img. However it seems we have to add water for a mix before using it.

[littleowl]

it is not even close as heavy as mercury. I have had it in my look for over a month and I have not seen any trouble yet.
I am trying to get Mcoffey to do some testing because martin is to busy.
I know I am not doing the work that needs to be done but you try and do some of this with out a thumb. :( it really is harder then you think. :shakes:

I diluted the mix because I was worried it was heavy like mercury but it is not and I would like to see it on my B3 quad


this is a note I got from the guy that sent this stuff to me.


There will be no issues with the nanofluid leaking. If it leaks with nanofluid that means it was leaking with water. It won't mess up any seals. It is no different from pumping water because the concentration is low enough to not affect the density...pumping liquid metal..ie..mercury..would destroy any pump. Nanofluids increase the convective heat transfer coefficient, which increases the amount of heat removed from the chip without affecting the temperatures - meaning you can put more heat into the chip because the nanofluid can get rid of it.

[relttem]

if you all don't read the rest of the forum there has been a good discussion here:

http://www.xtremesystems.org/forums/showthread.php?t=202442

[man_bear_pig]

Maximum coolant.

[relttem]

we should be getting test results any day now. We really have no idea what it is going to do in a WC system. Thru our lab tests we should a decent increase in the heat transfer ability of the nanofluid, but our flow was quite a bit higher. Therefore, our turbulent mixing was a good deal higher. Still, it should be interesting to see what it can do. As for the feel of it; it really feels like water..maybe with a tiny soapy feel to it.
------------------------------------------------------------------
3 regular posters on here have samples now..hopefully, we should get some info by the weekend. Tomshardware has a sample as well. And, so does a guy at Anandtech...did I say all of this already? We are still waiting to hear back from MaxPC. soon...

[relttem]

3 regular posters on here have samples now..hopefully, we should get some info by the weekend. Tomshardware has a sample as well. And, so does a guy at Anandtech...did I say all of this already? We are still waiting to hear back from MaxPC. soon...

[relttem]

first tests results are in by Skinnee..Littleowl did a little bit of testing..we are waiting for Bob

http://www.xtremesystems.org/forums/showthread.php?t=202442&page=5

[dopestuff]

thanks for the headsup relttem :up:
and interesting..

[Dangals]

Awesome work Skinnee - can't believe I missed those results - keep it up :up:

[skinnee]

I have a full write up coming, with some shots of the fluid and a UV reactive verdict. :D

I will post all of the info when I have some pics to go along with it. :up:

[leuler]

Confirms (for this forum) what has been found in other tests; nanofluids
transfer heat better than distilled water. The less turbulent the flow,
the greater the difference between nanofluids and water. That is why
the temp difference was more pronounced with low flow. A low restriction,
low to medium flow loop would see a nice improvement with a nanofluid.
A high restriction, high flow loop would not see too much improvement.
Why? Thermal conductivity of the cooling fluid matters more in heat transfer
when the flow is less turbulent and nanofluids conduct heat better. When
the flow is more turbulent, heat transfer due to convection is more important.

I'll give this link once again: http://www.electronics-cooling.com/html/2007_aug_a3.php

[relttem]

thanks for posting the link to that paper. If you go to MIT's mechanical engineering page they have some nice links in their nanofluid research center. They also have a nanofluid research center in the ChemE department, but the ME department is more relevant to heat transfer characteristics.
The nice thing about the link for the paper is that enough information is given to let you determine what kind of pumping implications the addition of copper nanoparticles will have. Again, you have to be careful in terms of the amount added, because it is a trade-off between the pumping power versus the enhancement. For example, you might max out the enhancement, but then your pump will only last a few months.