Solid Silver Waterblock?

[RoadWarrior]

I wonder about the directionality of thermal conductivity of diamond. You see a slice of diamond is gonna be along the plane of the crystal structure, and at a guess I'd say that's the plane of best conductivity, getting one that's cut in a thin slab right across the planes is gonna be difficult. Then I wonder, even if diamond is damn hard, would such a piece be brittle and hard to handle???

Actually, I keep hearing rumours of a crystalline form of gold being possible. It's not in the science books yet though. I would imagine some inconcievably high temperature and pressure would be necessary to accomplish this. Maybe we will find some on the inner planets.

Road Warrior

 

[futura2001]

If I buy a CVD diamond, it should solve that problem. CVDs are grown from a vapor. POCO foam is also a possibility, but I dislike the idea of using a foam substance for the contact of a block.
Futura

 

[RainMaQer]

I think the idea is good... in the fact that "man made" diamonds have excellent heat conductivity... but I think you'll run into some serious problems...
1: Water distubance... the diamond will be flat... VERY... even if you have it cut those cuts will still be smooth as heck. That will seriously degrade cooling as the water will just flow over it... not really grabbing any of the heat.
2: Cutting... even after you get the peice you may need to do some cutting... or even to try and roughen up the one side for water disturbance.

 

[Tecumseh]

Get a chunk of silver and have dangerden make you a silver Maze3. Even better yet try a copper silver alloy since its a little better than just silver. I'm shure you could work out a deal with them. They might even do it for free if you do a little article on the block once its finished.

Sorry, Bender. A copper-silver alloy is NOT better for heat
transfer than pure metal. The original guy that did the
test assumed that electron conduction directly predicted
heat conduction. He has recently stated that Cusil does
not have a higher thermal conductivity than copper or
silver.

 

[Tecumseh]

I wonder about the directionality of thermal conductivity of diamond.

Road Warrior

Diamond has perfect isotropic heat conduction in its solid
form. The conductivity is not planer like graphite.

 

[Tecumseh]

OK, are we all going off the deep end of the pool?

Experiment is fun, really fun, but we don't need anything
more exotic than what it takes to get the job done.

Synthetic diamond (bort) has been used for a long time.
This conductivity is not as good as your Mom's engagement
ring, but pretty good.

What is needed here is a System approach. Look at the
whole system at once. What are you trying to accomplish?
If the ultimate goal is overclocking, then spending money
on exotic methods which don't give you any more MHz
is not wise.

I think it prudent to have a complete design spec before
you spend the money. Just a suggestion.....

 

[SavageHenry]

Instead of diamond or silver, you might want to investigate using a heatpipe-based heat spreader. It would probably be cheapear and conduct heat better than silver or diamond.

Here's an page with some information: http://www.thermacore.com/thermabase.htm

 

[stevens]

Hey man i say go for it. I actually though about doing this but then i remembered i have no tool or money. I think that this could turn out to be really good looking and effective. To make turbulance on the diamond, take a drill and put little holes (that dont go all the way through obviously). You may want someone with a drill press to do this so that you would be able to get the same depth for all the little marks. and in the silver block , do a maze or a spiral pattern that starts in the center, and have the cool water come in from the center and cut out the center of the block under were the cool water comes in and that is were you should put the diamond, but then the water will still go all in the block thus removing any more excess heat.

 

[JFettig]

do you realize that DIAMOND IS THE HARDNESS OF 10!!!???

you cant drill thru it unless you have a diamond bit and that would prob cost more than the diamond!

 

[stevens]

dont drill through it, just make little indentations with the drill, and they do sell diamond tipped drill bits which are notvery expensive i dont think. All i am saying is that he need to make some sort of turbulance over the diamond to get good heat transfer. I think that little drill indents are a good way to do it. I am not sure about the drill bit though just asuming becuase i think i have seen them somewhere, but i may be wrong.

 

[Tecumseh]

Instead of diamond or silver, you might want to investigate using a heatpipe-based heat spreader. It would probably be cheapear and conduct heat better than silver or diamond.

Here's an page with some information: http://www.thermacore.com/thermabase.htm

You are on the right track! This is the future. Just remember
that heat pipes/vapor chambers can only move or spread
the heat efficiently. You still have to get rid of it.

 

[Tecumseh]

do you realize that DIAMOND IS THE HARDNESS OF 10!!!???

you cant drill thru it unless you have a diamond bit and that would prob cost more than the diamond!

Assuming you HAVE a diamond you want to use, it's very
easy to cut or shape it with a laser.

 

[SavageHenry]

Instead of diamond or silver, you might want to investigate using a heatpipe-based heat spreader. It would probably be cheapear and conduct heat better than silver or diamond.

Here's an page with some information: http://www.thermacore.com/thermabase.htm

 

[Aesik]

Instead of diamond or silver, you might want to investigate using a heatpipe-based heat spreader. It would probably be cheapear and conduct heat better than silver or diamond.

Here's an page with some information: http://www.thermacore.com/thermabase.htm

Unfortunately I think the idea is that he wants to design and build his own. Constructing working, effective heat pipes is not something you do on in your garage.

 

[futura2001]

The diamond still sounds possible, but I am having considerable amounts of trouble finding one online to purchase. Hopefully I will find one soon.
Futura

 

[RoadWarrior]

Unfortunately I think the idea is that he wants to design and build his own. Constructing working, effective heat pipes is not something you do on in your garage.

Yeah, who's got a garage these days? On the kitchen table with the dremel is more like it.

Seriously though, doing something with a half inch copper pipe, salvaged braid from the shield of thickwire coax for wicking, alcohol for refrigerant, ball valve on one end and a rebuilt bicycle pump and car cylinder pressure vacuum gauge, seems possible for intelligent men of an enterprising nature.

Road Warrior

 

[JFettig]

Yeah, who's got a garage these days? On the kitchen table with the dremel is more like it.

Seriously though, doing something with a half inch copper pipe, salvaged braid from the shield of thickwire coax for wicking, alcohol for refrigerant, ball valve on one end and a rebuilt bicycle pump and car cylinder pressure vacuum gauge, seems possible for intelligent men of an enterprising nature.

Road Warrior

im gonna make a heatpipe when i get a job, just use r134a.. and do the valves at top and bottom and find the rite amount...

 

[futura2001]

I am going to post a pic of the design for the block tomorrow if I can. I have decided to use a block of copper and a plate of pure silver with a diamond embedded in the silver via silver solder. To create the maximum heat flow, I hadn't taken the smooth diamond surface into account, I will drop tiny beads of solder in a hex layout all over the silver plate and diamond if I cannot "dimple" the diamond with a drill. Over that will be a modified metal fan. Instead of the smooth blades like normal fans have, I will file tiny notches to add to the micro eddy effect and increase turbulance. The fan will be spun by two jets of water intakes and two exhausts. That should net me the max amount of turbulance possible...
Futura

 

[Malpine Walis]

What about the thermal grease?

While the idea of a diamond waterblock is interesting, nobody has yet mentioned how the heat is going to get from the core to the diamond

Now the diamond may have a very smooth surface but there aint no such thing as perfectly flat! Since the only way the suface can be lapped is with diamond dust, the particle size that was used for the lapping will determine the smoothness of the surface. You are going to have to put something under your diamond to assure thermal transfer.

The best product on the matket today is probably arctic silver 3. Which works wonders under a piece of copper because, in part, silver is a better conductor than copper. But under a diamond? I would think that the diamond would be a waste if it can only get as much heat off the core as AS3 will conduct.

If you are really going to do this, you may need to develop your own version of thermal grease. Perhaps the folks who make AS3 would be willing to help. After all, if it works, they can start selling Arctic Diamond

 

[res0r9lm]

arctic silver tried using diamond paste but didn't work too good. had something to do with molecules not meshing together as well as silver. I think ocz is trying to push a diamond paste.