Arctic Silver 5

[cjolley]

Gold leaf?
chuck

[Greebe]

What the heck?! Why on earth would anyone want to use Lead or Tin/Lead mix as a thermal conductor. It is truely a poor thermal conductor... even gold leaf is only half as good as Copper and 2/3 as good as Al.

The only elemental material that exceeds copper is pure silver... and I'm not refering to Sterling, for to be classified as that it only needs to be =/>72% pure

[Brian R.]

The difference between thermal properties of a thin film of different pure metals is miniscule because it is a function of thickness. The thinner, the less difference a difference in thermal conductivity would make.

Chuck, Gold leaf is probably a better choice. Thinner than the typical lead foil I've seen. Which one is more maleable? That would be the deciding property I guess...

I think gold is more maleable...

[Greebe]

@Admiral, please heed, you are being over zealous about this. (and being told by an EE). What you are doing is fine, but you will never see the full benefits of using any HS with sillycone grease

[Greebe]

@Brian R. *sigh*. If you wish to counter my claims I would first suggest you go back to school and become an EE or better yet a degree in physics specializing in thermal dynamics for you would never made that coment if truely knew what you are saying.

[Brian R.]

Yes Boss, anything you say. pfft

[Wombat]

Personally, I prefer keeping the computer in the ocean. The salt water absorbs a ton of heat, and it's got amazing flow rate.

[cjolley]

Originally posted by Wombat
Personally, I prefer keeping the computer in the ocean. The salt water absorbs a ton of heat, and it's got amazing flow rate.

Remember that guy who put one in refrigerated mineral oil?
I think it was a 486 so it's been a while.
chuck

PS, Here's a nice practical solution http://www.muropaketti.com/artikkeli...0/index3.phtml

[Brian R.]

Well shit, if the Boss won't let me talk about metals, maybe I can talk about oils then.

I would think mineral oil was a bad choice since it gets very viscous when cold. The heat transfer in the cooling coils would be alot slower than you want. Pumping it out of the cooling coils would be a problem too. I think one of these new 0W-20 synthetic motor oils might be alot better than mineral oil.

[cjolley]

Originally posted by Greebe
... even gold leaf is only half as good as Copper and 2/3 as good as Al...

Hey Greebe,
I thought gold was a better conductor than copper.
Does it's Thermal conductivity not follow it's Electrical conductivity exactly?
Or am I not remembering it's electrical conductivity correctly.
I realy didn't think gold (even leaf) would realy be soft enough.
chuck

PS Oddly enough, I dont have my ~1980 MERC INDEX with me at work now that I haven't been teaching for 20 years

PPS I've about had emough of my spelling

[Brian R.]

Thermal conductivities at 25 C:
Aluminum 2.36
Copper 4.01
Gold 3.18
Silver 4.28

Gee, I hope I don't have to be a EE to say this...

But the units are Watt/cmK

Note that the units have cm in the denominator which means the conductivity is inversely proportional to the thickness. Like I said above (EEs please take note)

Edit to correct data

[Jon P. Inghram]

Two different solutions to this problem: Cold welding, or put an amorphous diamond coating on the chip and sink and make sure it's polished flat enough as to not need any transfer medium.

[Greebe]

I find it rather amusing when anyone who goes outside of their respective field counters one whom is stating facts from their respective field

Must be that chip on your shoulder.

[Brian R.]

No, I have a head on my shoulders, unlike people who rely only on education

And if above, you were trying to make the argument that the surface conductivities were inadequate using a metal foil then just say so. That wasn't what I said and I am well aware of this deficiency.

[Jon P. Inghram]

That's not right, Silver is is a better thermal and electrical conductor.

At 26.85c (300k)

Silver: 0.429 watts/millimeter/degree kelvin (or celsius, they're the same size)

Copper: 0.401 watts/millimeter/degree kelvin