Carbon, graphite, graphite foam block and rad ideas

[XS Janus]

OK, since Dinos22 nicked the subject in the other thread ( http://www.xtremesystems.org/forums/showthread.php?t=137493 )I'm posting this here.
He asked if the whole carbon cooler thing could be made to work in WC setups.
My intention by this thread is to get some brainstorming going on this subject to see where it will lead us.
Last 2 days I was browsing the net looking for info on Graphite blocks and foams and stuff.
I have downloaded quite a bit of formal pdfs and links with lots of interesting stuff to learn.
Since it is a lot to read and English is not my language, It will take me a while to absorb all the stuff that could be use full to us.
I uploaded all the files in one 13,5 MB rar archive on rapidshare.
http://rapidshare.com/files/22463843/Graphite_and_G_Foam_links_and_PDFs.rar.html
Anybody who is interested please download and at least glance over it.
"Serious folks" are considering all these concepts and are exited, but the process is slow, but we can build and play ourselves.:)
The idea was done before of course (LONG before actually yr 2001). by this guy: http://www.spodesabode.com/content/article/poco
He used foam named Pocofoam in his DIY block and while his method of implementation was somewhat crude by todays standards, but it was a cool idea that never caught on. pics are below...
The sample he used is today even better and there are 2 versions of the stuff on pocographite site.
That same foam is a hard substance and can be machined as desired.
The whole concept is even more envisioned by other industries as using the GF instead of fins on heat exchangers, radiators or regular heat sinks.
Funny thing is they also have a online store: http://www.pocographiteonline.com/servlet/StoreFront
Their web page is www.poco.com

Other graphite and carbon products can be found for example here:
http://www.graphitestore.com/index.asp no foam though.

Any info numbers and links feel free to contribute here.
Sorry for the long post.
Remember let's keep a positive note in this thread, please not a lot of posts how this and that won't work but rather how and why to make it work.

Thanks guys-gals for all your ideas and input you put into this thread and for your time in participating.
I'm sure we'll make something of it, it is XtremeSystems after all if we can't do it nobody can! :D
XS Janus

[XS Janus]

Oh come on, I can't believe nobody has an opinion on the stuff!
I was actually hoping some cooling guru tried it before or knew more about it. I guess nobody has some valuable info to transfer to us minions...
Or people gotten just like to buying finished products to much I guess.:(

Here are some very simple concepts pulled out from one of the studies in pdf

[mcoffey]

OK,

I'll give you one. I don't really understand it, and it looks butt nasty IMO.

Never considered graphite as a good conductor of heat. Not saying it wouldn't work, just not very attractive based on what you've show here. and I'd be concered with grit and grime from the graphite in the loop.

and unless it's some super conducter of heat heads and shoulders above copper, it's way expensive, and I quote "They replied that they were just entering production and unfortunately the 2"x2"x2" engineering sample I requested would cost around $480 usd.". The 1/4 inch sample he used was 200.00 bucks. AT that price, I'd be expecting a miraculous improvement, and I don;t see it based on the lit provided.

and I'd be concerned about the pourous surface becoming pluged up. If people are having to clean out the Storm jet every other month, and that opening's at least 10 times larger than that of the pourous surface, can't imagine it working well. The graphite block would become pluged up over time and rendered useless IMO.

Sorry, you asked.

andyc

[dinos22]

LOL someone reads my posts and takes them seriously hehehe (i don't know i nicked anything though but just realised that you were posting a fair bit in the thread about this....i didn't read it and lots of my posts have been deleted....oversensitive mods hehe) anyways.............

it would be interesting to hear from Cathar and some of the other block designers on this idea as i'm sure they would have experimented with the idea at some point :)

good on you for doing so much background research

one thing i don't doubt is the fact it would work as carbon with "microchannels" or whatever they called it in that thread is used to absorb that heat and send it to the heatpipes so i guess with a decent design water could cool the absorbed heat as well.................

would be interesting if Gabe has anything on this as well.........Swiftech is very switched on with new ideas and think progressively :)

i think you should also link to the original thread in which the carbon air cooler from OCZ is mentioned........just to give the background info on this subject :toast:

[welshtom]

Diamond being the best heat conductor out there, i think that this idea has great potential.

Carbon bottomed waterblocks would be better than copper IF you can get the heat away from the carbon efficiently as copper transfers it to the water.

Tom

[Pete]

I'll download soon as i finshed sorting this laptop out and have a read!

As the big man aboiut said. Diamonds are infact a form of carbon. I think it you were to do blend of the carbon foam with copper you might be on to a winner. I know OCZ have some air cooler thing which is big and inslightly coming out!

I look forward to the read

[Locke]

What about copper plated carbon? you could machine the carbon block into any shape, with fins or holes for water flow to go through, with the copper plating shielding any chips or grit from entering the system.

also, synthetic diamond might not be that expensive in thin sheets and possibly could be bonded to a copper block, however it would be very brittle.

also, silver is a better conductor than copper, it just tarnishes alot faster... the nice thing about silver is it can be cast and is relatively cheap (14$ an ounce...) I wonder if clear coating with acrylic would inhibit thermal transfer...

[[XC] gomeler]

Coating the silver with acrylic would kill the ability to transfer heat. I've thought about this a bit and if the brittle tendancies of the graphite can be canceled out then it would make an amazing conductor. Diamond has a thermal transfer that is 5 times that of copper so graphite should be pretty close to that, atleast better than copper. OCZ's use of graphite will be very intersting to see, we might see a change in the industry from copper to graphic on the CPU interfaces.

[Locke]

Yeah this is a fascinating materials problem, which i bet the answer for is simple... Ill certainly put some thought into it and ask some industy people I run into along the way. HMM maybe ill start pricing zirconias lol

[Fr3ak]

As long as the IHS is made out of copper and standard tim being used, I dont think even that superconductor will be as good as you hope it would be.

Tim on the IHS and the IHS are 2 things that have to be improved in order to get way better temps and faster heattransfer. I dont think improvements on the waterblock itself will gain more than a couple degrees, if any at all.

[serialk11r]

Graphite is different from diamond, I hope you knew that pete.
The diamond block thing, well, theres quite a few people here (you know who you are ;)) who have a ton of money to spend, hehehe maybe you guys can get a waterblock made out of synthetic diamond?

[Allsorts]

Yeah, even though graphite and diamond are both forms of pure carbon they have wildly different physical properties. Graphite does have a slightly higher specific heat capacity than both copper and diamond but it's thermal conductivity is quite frankly pants. Having a porous structure sounds interesting though; more surface area to transfer heat.

[serialk11r]

Yeah, even though graphite and diamond are both forms of pure carbon they have wildly different physical properties. Graphite does have a slightly higher specific heat capacity than both copper and diamond but it's thermal conductivity is quite frankly pants. Having a porous structure sounds interesting though; more surface area to transfer heat.

Well for a waterblock you want low specific heat capacity...

[XS Janus]

Thanks for the replies guys and gals, must have been the weekend dead zone when I first posted this:D

From the manufacturers site you can see they now have 2 types of foam: normal and one named HTC. The new one (Poco HTC) was produced for the reason of better thermal management and better machining properties.
From what I understood they have a procedure and "grow" the foam in one direction that stretches the and form a continuous line that go around the pores.

This gives the foam its great T-conductivity in one direction. The POCO HTC version is much harder and withstands higher loads on top of it, meaning it's denser, heavier, more heat conductive, but less porous and you can machine fins in it.
That said my guess is that if one would try to force watter though it (like a sponge) it would be a serious flow killer(read stopper) and would require to much pressure and to strong of a pump to justify it.

I read that they are experimenting with copper and metal plating for the contact surface area, but on their web they are advising the use of S-bond to fill in the pores on the contact surface and to bond Poco with other surface. www.s-bond.com is the site mentioned.
Haven't had the time yet but I expect that to be some kind of a thermal epoxy or something simmilar

Being that the foam conducts heat best against the plane maybe fin designs that would maximise vertical surface would work best.
They also are doing research of using the foam as a substitute for aluminium fins in radiators where the foam is placed directly on the copper piping or other type of heat pipe design and air is pushed on it or through it at some degree.

I have collected some other links+files and will post a rapid-share link when i sort those out and get away from daily chores:D

Here are a few links that might give some more inside in the whole thing
http://www.pocothermalcs.com/faq.asp Answered questions people asked
http://www.poco.com/us/Thermal/htc.asp POCO HTC technical info, pressure drop and other graphs
http://www.poco.com/us/Thermal/machining.asp Machining info
http://www.poco.com/us/Thermal/applications.asp Some Application they people are thinking up

[Fossil]

The properties of this foam stuff are probably very different from ordinary graphite.

It would be great if we could get some kind of mounting material that will conform to give good contact all over a dodgy, non-flat IHS.

Maybe this stuff wouldn't help with that, but it's something to wish for.

[XS Janus]

Soccerswim88, you are somewhat right I'm still not sure as to why this would work better than some of todays blocks, myself but can't seem to shake the idea that it is just a matter of implementing it right and using it so its strong sides come to play even more. We know that a lot of stuff goes into how a meterial works in this conditions, for instance how well it "gives away" heat and what grade of the material is used in normal heatsinks.

http://www.ornl.gov/info/ornlreview/v33_3_00/foam.htm Inventor talking

http://www.osti.gov/energycitations/product.biblio.jsp?osti_id=885604
It was shown that by corrugating the foam parallel to fluid flow, the pressure drop can be reduced significantly while maintaining the same heat transfer as that in the folded fin heat sink. In fact, the results show that the graphite foam heat sink can utilized 5% the pumping power as that required with the folded fin aluminum heat sink, yet remove the same amount of heat. -What ever the las sentence means...

A post on by James Klett ORNL official in on one forum said this:
"Basically, we have made heat sinks for computers that beat the compitition by 30% at the same air flow. I have publications on it that I can share. There are even liquid cooling techniques and vapor cooling techniques that blow away the competition. I could even teach you how to do it on your home computer. We have demonstrated power densities over 150W/cm2 at chip temps of 70C. BTW - the shuttle only sees 40W/cm2 when it re-enters the earths atm. It is all very simple, but may be costly.