Strange idea I got while I was drunk - Please Read!

[Intraveinous]

OK, so this is probably a really dumb idea as I was drunk when I thought it up, but I figgered I'd throw it past the PhDs before I went out and spent a bunch of money on something that would never work anyway... So here's my idea...

A Glass Waterblock

I'm no expert in thermal dynamics, but there are a few properties of glass that I think would make it a good candidate, and several that help to disqualify it...

Anyway first the pros:
Pro 1. It seems to dissapate heat to water really well without holding much of it. Glass cookware, boiling water, cold handle. That could go both ways, that it doesn't transfer heat through itself very well, but I dunno, that's why I'm posting here...

Pro 2. it'd look f#*(#n sweet to have a completely clear block and be able to see your chip under a swirl of water while it was running, even better if you have some kind of dye in your water and a Blacklight...

Con 1. Harder than hell to machine, next to impossible to mass produce.

Con 2. Fragile, both for mounting systems, and for transportation.

So yeah, there are other pros and cons of course, but those are the ones that come to mind first for me anyways... So yeah, somebody tell me I'm a dumbass and why, or somebody tell me I'm a genius and I'll get on making one...
Peace
John

[nihili]

Well the first thing to do is to research the thermal properties of glass. While you're at it, check lead crystal too. I suspect that it doesn't transfer heat nearly as well as metals do, but I don't know.

As for seeing your cpu, it would probably be obscured by the thermal grease. But you could still see the water. Of course if you just want to see the water, then you can just put a glass top on a copper block.

I don't see a problem with production. People have been producing cut glass for years. Also, you could probably do blown glass. Remember a lot of very precise lab implements are made from glass.

nihili

[William]

Glass isn't going to transfer the heat very well, but this would definately work with a copper or aluminum bottom to a glass block. It would probably be relatively cheap and pretty strong if you used the right kind of glass.

Unlike Nihili, I don't have a Ph.D :rolleyes:

[nihili]

Guys, the PhD is irrelevant. And I mean that in every possible way.

nihili

[Silversinksam]

Despite William not having a PHd he is correct, glass is a horrible conduit, heres the list:


Thermal Properties of Materials
Thermal Conductivity, W/cm-K
Metals
Aluminum 2.165
Beryllium 1.772
Beryllium-copper 1.063
Brass 70% copper, 30% zinc 1.220
Copper 3.937
Gold 2.913
Iron .669
Lead .343
Magnesium 1.575
Molybdenum 1.299
Monel .197
Nickel .906
Platinum .734
Silver 4.173
Stainless Steel-321 .146
Stainless Steel-410 .240
Steel, low carbon .669
Tin .630
Titanium .157
Tungsten 1.969
Zinc 1.024
Semiconductors
GaAs .591
Silicon (pure) 1.457
Silicon (.0025 ohm-cm) .984

Silicon Dioxide (amorphous) .014
Silicon Dioxide (quartz) c-axis .11
Silicon Dioxide (quartz) a-axis .059
Silicon Nitride .16 - .33
Silicon Carbide .90
Insulators
Air (still) .0003
Sapphire c-axis .35
Sapphire a-axis .32
Alumina .276
Alumina 85% .118
Beryllia 99.5% 1.969
Beryllia 97% 1.575
Beryllia 95% 1.161
Boron Nitride (hot pressed) .394
Diamond (room temperature) 6.299 :)
Diamond (77 K) 24.
Diamond (room temperature, isotopically pure) 50.
Epoxy .002
Thermally conductive epoxy .008
Glass .008 :eek:
Heat sink compound (metal oxide loaded grease) .004
Mica .007
Mylar .002
Phenolic .002
Silicone Grease .002
Silicone Rubber .002
Teflon .002
FR-4 or G-10 PC board material .003
water .0055
Liquid Helium (4.2K) .000307
Liquid Nitrogen (77K) .001411
Liquid Argon (85K) .001258
Thermally Conductive Elastomers
Bergquist Sil-pads .009
Tecknit Consil-C 871 .023
Tecknit Consil-R 350 .00433 to .00732
Saracon 2.9e-3 cal/cm-sec-K
Chomerics XTS-274 alumina filled elastomer .002 cal/sec cm K
Cho-seal 1224 .038
Cho-therm .0433
Cho-therm 1678 .018
Cho-therm 1671 .027

[IFMU]

Ok well I cant get the dang right link to post here so I bumped it... But This is the easiest way to do what your talkin about... so its at the top of the cooling right below this one... Its started by me and its called Home Built Waterblock...

[ButcherUK]

hmmm so how much is isotopically pure diamond... 50W/cm-K is pretty damn good and being diamond it's transparent :D

[Intraveinous]

Well of course, Isotopically pure diamond, I've had 50 pounds of the stuff sitting out in the garage and haven't known what to do with it :D
Well, as I thought, it's not a great idea, so I'm glad I put it here before going and getting glass cutting implements. As as far as machining/mass production, I meant for a DIY project typa thing... As far as PhDs, it was figurative, though I figured nihili would respond :D. Thanks a lot for all that info Silversinksam!
Peace
John

[SavageHenry]

Instead of a glass top for a copper waterblock, how about a Lexan top? It's lighter, less brittle (IIRC), and it can have threads tapped into it for hose fittings to screw into.

[nihili]

Silversinksam,

I remember my mom used to have a fancy frying pan that had a hollow bottom filled with oil. Any idea what the thermal properties of oil, alcohol, etc are?

nihili

[lennytiger]

Oil can be processed into many different forms, its viscosity is very important crude oil i think has a high viscosity i'm not sure this is correct but vegetable oil conducts heat very well because it is not a very dense oil. Vegetable oil frys steaks though...

I have no Phd or any kind of a qualification as of yet, so don't trust me. Please correct me if I'm wrong!!


Somewhere on the frontpage, there is the silentium project which uses oil cooling, try looking there.

[nihili]

Originally posted by lennytiger
Oil can be processed into many different forms, its viscosity is very important crude oil i think has a high viscosity i'm not sure this is correct but vegetable oil conducts heat very well because it is not a very dense oil. Vegetable oil frys steaks though...

I have no Phd or any kind of a qualification as of yet, so don't trust me. Please correct me if I'm wrong!!


Somewhere on the frontpage, there is the silentium project which uses oil cooling, try looking there.

Thanks, I'll have a look.

I've got a PhD and still don't have any qualifications

nihili

[ButcherUK]

I read someone about someone having good results using precision instruments oil - it's very thin to allow it to work into small cracks easily.

[lennytiger]

ButcherUK Welcome to the forums,

This can also be acheived by passing low viscosity thermal grease through a watercooling system.

Thermal grease is a conductive liquid and is used on the stock CPUs Intel AND AMD Supply!

[Pepsi]

Funny i had the same idea I'll let you know how it works out.

[IFMU]

Ok, seeing a name like that - Pepsi - I just had to jump in and say Welcome aboard and LOVE the name.. well ok actually more the drink, but all technicallities... :p

[Maximus Nickus]

Glass waterblock? You might as well have a paper bath!:D :D :D

[Intraveinous]

Originally posted by nick_cw
Glass waterblock? You might as well have a paper bath!:D :D :D
Paper bath??? Am I missing something or am I just dumb today??? I kinda figured it was a crackheaded pipedream, but I had to run it by here first to be sure.
Peace
John

[Wa11y]

Originally posted by nihili
Thanks, I'll have a look.

I've got a PhD and still don't have any qualifications

nihili

You have a PhD? In what? And can I borrow it sometime? Women are impressed by doctors.

[Nhut Pham]

nihili, who exactly are you? I'm new to this forum, and I'm a freshman attending U of I in urbana champaign. I was wondering if you're a student or maybe a teacher here? Or do you just live in champaign? You make me very curious, you can contact me at nhutpham@uiuc.edu

[nihili]

I'm a professor of philosophy. I specialize in Logic.
My uiuc email is jimhardy@uiuc.edu

nihili