Direct Die phase change cooling?

Well, I have been doing a bit of thinking, mainly along the lines of bonding a chip directly as a part of the expansion chamber. Now I realize that this would be a bit difficult as it would have to withstand high high pressures, but is this possible and has it been done?
I am fully reticent of the fact that I am exposing a seal to ambient temps on one side and temperatures well into the -40s on the other side.
Still, there should be something that would work for this. I am sure there is something rated for high PSIs.

Otherwise, I am going to have to get back to my main project and that means I have to mill a block and I am trying to find something to distract me from that slight unpleasantness for a while. Like DD phase change.
Futura

well...you could do direct die with a block similar to bowman....and the freon or whatever u decide to use doesn't actually touch the chip but the bottom of the copper block....but again that would involve milling...something you apparently don't want to do.

I had a similar thought about this today as well but im not entirely sure the core itself would able to contain or withstand such high pressures....it is only silicon...(i think)...Also, it would be difficult to use a different cpu if you ever wanted to switch it out with something.

yeah the core wouldnt be able to take 150 psi. Now if you cooled down the condensor the psi might drop to 60 or so and that might work. I would just make a block, its a ton easier.

ha ha you guys though i was dead......


there are some problems with direct die cooling.i have thought about it...hummmm. well the thing that worried me the most beside the leaking...was the spray of refrigerant.it you dont get a even spray pattern your chip will heat in small pockets,and i believe burn up.unlike water it is impossable to see the spray patten or where it will be hitting.it would boil off so quick i dont think it would ever get a chance to puddle up over the core.

you could flood it. Also you can make an incloser thats not much bigger than the core so there is little room for the refrigerant to go besides on the core.

Yes creating the block would be easy enough, the materials I would chose would be lexan and copper foil.
Lexan should take alot of pressure, and the copper foil could be used to make either a maze/spiral design or just an enclosure over the cpu core itself.Then I would another layer of lexan around the block with a vacuum between them so it would be insulated and you could see the refrigerant(looks pretty damn cool).

But the cpu probably can't handle the pressure, unless of course you wanted to make a cascade system

Hmmm...
I might want to contact AMD about how many PSIs the core can stand. (A response of most likely 'Huh?') But this is intriguing. I could just get a duron first and test it out on that. I can definitely see some major problems with this, but if it were to work, it would be damn interesting.
Another thing to consider is whether the chip itself provides enough surface area to even make this viable?
My main qualm with milling a block is the fact that I am going to have to hollow out a solid piece of lexan to hold the copper piece in, either way I cool, I am going to have to eventually have to do this, but at least this way I can prolong the inevitable a few more days.
Futura

Maybe a better way will be to make box around the chip with socket in it?
In this way the seal who’d be not on the chip but on the socket.

I"ve thought about this for countless hours.

If you can get the seal to hold in a deep vaccum with out damaging your m/b or chip, your atleast on the right track. I can only see this working if you use a open copper box with that Epoxy Puddy and Neopr as the seal. Now keeping the seal at sub0 temps is were the real trick somes to play. With all the different materials I can see alot of expantion and contraction going on at different rates.

All the common Refrigerants are very stable and compadable with almost everthing. The problem is POE oil. It doesn't seem to be stable with anything other than metal. So you would have to go with Alky oil and a HCFC type refrigerant like MP-66. Mineral and using R-22 would work, I wouldnt' suggest using Propane or any other Butane type for such an application.

1 you will have to be careful in witch refrigerant you use that it will not be a conductor of elect.
2 No make the box on a separate socket that will plug in the motherboard socket.

Just in case you didn't know all the Normaly used Refrigerants do not conduct at all. Something possibly closer to home... all the original Halon system were just refrigerant (R-114 I think). Seeing they displace the oxygen in the room and are prefectly safe on all Electronics.. That Electronics Spray Cleaner we used to buy back in the 80s.. that had like 50% R-22 in it.

cool