Absolute best heat spreader,short of diamond, i found for sale!

Absolute best heat spreader,short of diamond, just insanely expensive!

Sadly it costs way way to much for most anybody other than nasa or people with really high end equipment. Hopefully they drop the price ALOT soon.

http://www.2spi.com/catalog/new/hopgsub.shtml

If anybody has dealt with these, know a much lower price etc please post or PM me as i really would like this on a heatsink or waterblock. I hope this is interesting and sombody knows of this for less than $1,000~$3,000

3000!!!

Since graphite is pure carbon, it is, in a sense, diamond, since both are carbon, just in different crystallized forms.

Yeah, but graphite and diamond have entirely different characteristics. Diamond is hard, graphite is soft. There is no reason to assume that their heat conductivities will be the same, either. Chemical behavior depends a lot more on the structure of something than on it's atoms.
It's also notable that this heat conductivity is only high ALONG the lengths of the columns. It is low perpendicualr to that. So if you put a block of this on your processor, it would not spread the heat very much, it would just transfer it straight out, leaving you as much of a problem to cool it at the other end.

Well ya... but thats the same for a heatsink of any material... thats why they make em with large surface areas so the heat is spread and dissapated much easier. If you made a good heatsink out of it im sure it would be awsome.. but still a waste of money.

This is why they are working on organic processors, and processors that make no heat whatsoever, such as optical CPUs.

Actually, a heatsink made of this would suck... heat would mostly go out the tops of the fins directly over the core. This does not transfer heat equally in all directions like metal does. Here is a picture. The top shows how heat is spread in copper. The bottom shows how heat would be spread in this material.

Yeah, HOPG might be a nice material, but a coldplate 50mm x 50 mm x 4 mm of the lowest quality costs !! 2100.00 !! and thats dollars. Oh well, if we join forces we can by ten of 'em for just $1890 a piece !!! LOL!

Yep, utter madness....

I have a slab of PG, pyrolytic graphite. It probably isn't "high ordered". I have yet to test if it carries heat in all three directions.
I made a waterblock with it, but hasn't really compared it to others... It has a heat conductive capability of around 1300-1800, where copper has 380 or something. I don't remember the correct term for it at the moment....

You can use HOPG for heatsinks/waterblocks, you just have to think a bit more about the heat path. I still have tyo find out if it can beat copper WB's even though it might be a bit handicapped by the 2D heatpath.

hmm if the hopg has such a 2d heat spreading path....then chances are its used in a VERY thin layer. Like from what i reread the site, you cleave the 1mm layers into 10 layers or such to do whatever they do with hopg nowdays. Hopg is Highly Organized Pyrolytic Graphite.
So i would have thought the heat spreading would be much better as from what i have read. For that graphic on the heat spreading what size is the heat source and wattage? That graphic still could pretty much show Hopg is 3~4x better depeding on heat source,size etc.

wildfrogman said:

Hopg is Highly Organized Pyrolytic Graphite.

Click to expand...

No, HOPG is highly ORDERED pyrolytic graphite.
We overclockers cant use any PG in a thin slab of 2 or 1mm, it's a very fragile material.
The PG I had could sort of be treated as the "lead" in lead pencils.
As fragile, and when machined, it makes dust that discolors all equipment. It's a real challenge to work with, b'coz of the frailty.
Drilling is hopeless, all exitholes are chipped around from the drill going through, even though I drilled with the mill, set on slow slow slow feed/move. The chipping is also evident when you are milling it flat and comes to a corner/end. You always have to make sure the "cuts" are made "in towards the center". The material can be milled fast, sanding is easy/fast.

As for results with my home made PG waterblock(possibly a crappy design), well, all my data contains so many uncertainties. I cannot guarantee/warrant any of theese figures. LOL!

But if you want some soft data:
Pump: Hydor LE30, System flow about 6L/minute obtained by "filling the bucket/time"
2 Rads (made for freon cooling size 20 x 20 x 5 cm or 8 x 8 x 2 inch.)
P3 750 running at 1,9 volts oc 930 (7,5 x 124)

Water temps 25 (a stripped down in/outdoor home tempmeter, ie a peice of crap)
On die diode temp 32 (as per Asus P3V-E 810i chipset)
Ambient temp 20 - 21 (the same in/outdoor crap)

Well, I have no idea as of any precision in any of theese measurements. It doesn't look to bad, but it sure as hell doesn't look revolutionary.

hmm, thats what i get for posting what acronyms stand for off the top of my head. Ohwell, yes to check back again that does make sense. I am not sure but i think i have seen where this PG or such was either plated or covered with a special ceramic plate and somhow sealed on very well so its protected from damage. That seems to make sense considering that the carbon is very fragile.

Yeah... it should perform slightly better than copper, but slightly worse than direct die.... Go figure. A better design might be the best thing short of direct die, but I doubt any passive cooling solution can beat direct die watercooling.