• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

As Good As Gold

BonzaiDuck

BonzaiDuck

Lifer
I happened across an archived project posted (I think) at OverClockers.com.

The poster had augmented his CPU HSF with a commemorative silver dollar, and posted some data on the the thermal resistance of substances in the Periodic Table of Elements.

He had lapped the coin's finish until it was smooth, bent the clips for his HSF slightly to accommodate the additional 1/8" thickness of the silver slug.

The lowest thermal resistance is provided by the following substances in ascending order (higher thermal resistance from top to bottom in the list🙂

Diamond
Gold
Silver
Copper
Aluminum

Thus the choice of silver in Arctic Silver thermal compound.

I was watching late-night TV the other night, and they were offering a commemorative 20-dollar gold-piece last minted during the Roosevelt Presidency. They were offering it for something like $37 through a 1-800 phone-number.

The Golden Eagle Tribute (?) coin is gold plated. It would seem that this would be all that's required to lower the thermal resistance of a heatpipe cooler like the XP-120 -- it might possibly be better than the recently released SI-120, and it would certainly improve the thermal resistance of the latter.

Comments or observations? Caller, You're on the Air . . . .
 
Thermal Conductivity: W/(cm x K) @ 1 atm, 25 degrees Celsius

Silver = 4.29
Copper = 4.01
Gold = 3.17
Aluminum = 2.37
Platinum = .716
Mercury = .0834

So I'm not quite sure as to how your post reads...
Tas.
 
Originally posted by: tasburrfoot78362
Thermal Conductivity: W/(cm x K) @ 1 atm, 25 degrees Celsius

Silver = 4.29
Copper = 4.01
Gold = 3.17
Aluminum = 2.37
Platinum = .716
Mercury = .0834

So I'm not quite sure as to how your post reads...
Tas.
Click to expand...

Thats what I thought with Diamond being numero uno.
 
Originally posted by: BonzaiDuck
I happened across an archived project posted (I think) at OverClockers.com.

The poster had augmented his CPU HSF with a commemorative silver dollar, and posted some data on the the thermal resistance of substances in the Periodic Table of Elements.

He had lapped the coin's finish until it was smooth, bent the clips for his HSF slightly to accommodate the additional 1/8" thickness of the silver slug.

The lowest thermal resistance is provided by the following substances in ascending order (higher thermal resistance from top to bottom in the list🙂

Diamond
Gold
Silver
Copper
Aluminum

Thus the choice of silver in Arctic Silver thermal compound.

I was watching late-night TV the other night, and they were offering a commemorative 20-dollar gold-piece last minted during the Roosevelt Presidency. They were offering it for something like $37 through a 1-800 phone-number.

The Golden Eagle Tribute (?) coin is gold plated. It would seem that this would be all that's required to lower the thermal resistance of a heatpipe cooler like the XP-120 -- it might possibly be better than the recently released SI-120, and it would certainly improve the thermal resistance of the latter.

Comments or observations? Caller, You're on the Air . . . .
Click to expand...

I think I'll just grab my moms diamond bracelet, melt it down, make a wafer thin layer to the exact measurements of an AMD heatspreader and lay it inbetween my XP120 and the heatspreader. ftw!🙂 No seriously, I would love to see a silver based HS come out, but you know it would be $$. I thought I read somewhere that there was such a thing 😕
EDIT: Ahhh yes....Akaska came out with one...The Silver Mountain.
 
Originally posted by: BonzaiDuck
The Golden Eagle Tribute (?) coin is gold plated. It would seem that this would be all that's required to lower the thermal resistance of a heatpipe cooler like the XP-120 -- it might possibly be better than the recently released SI-120, and it would certainly improve the thermal resistance of the latter.
Click to expand...

Also, going with something that is just plated would not be the best thing to do. Even if you were to use it, you are still only conducting heat via the plated area... And if you had a bad material in the middle, such as Zinc (thermal conductivity of 1.16), it will act as an insulator, and it will cause that gold to get mighty hot, possibly causing it to come off. And that would be very bad.
Tas.
 
Also, going with something that is just plated would not be the best thing to do. Even if you were to use it, you are still only conducting heat via the plated area... And if you had a bad material in the middle, such as Zinc (thermal conductivity of 1.16), it will act as an insulator, and it will cause that gold to get mighty hot, possibly causing it to come off. And that would be very bad.
Tas.

So could we say that lapping the bottom of heatsinks was also not very good?
 
Originally posted by: JBDan
I think I'll just grab my moms diamond bracelet, melt it down, make a wafer thin layer to the exact measurements of an AMD heatspreader and lay it inbetween my XP120 and the heatspreader. ftw!🙂 No seriously, I would love to see a silver based HS come out, but you know it would be $$. I thought I read somewhere that there was such a thing 😕
EDIT: Ahhh yes....Akaska came out with one...The Silver Mountain.
Click to expand...

Higher thermal resistance = bad. You want thermal conductance, which is the ability to conduct heat. You'd be better off cutting a disc out of rubber matting and using it, as it won't matter to your processor for very long that you used rubber instead of diamond.
Tas.
 
Originally posted by: Zarubable
Also, going with something that is just plated would not be the best thing to do. Even if you were to use it, you are still only conducting heat via the plated area... And if you had a bad material in the middle, such as Zinc (thermal conductivity of 1.16), it will act as an insulator, and it will cause that gold to get mighty hot, possibly causing it to come off. And that would be very bad.
Tas.

So could we say that lapping the bottom of heatsinks was also not very good?
Click to expand...

Not necessarily. Lapping removes *tiny* amounts of material to even the surface. Also, for a heat sink, they are either all copper, or copper coated / plated aluminum. In the case of coins, such as the case with the penny, it is copper coated zinc. Aluminum is a much better conductor of heat than zinc.
Tas.

 
Originally posted by: tasburrfoot78362
Originally posted by: JBDan
I think I'll just grab my moms diamond bracelet, melt it down, make a wafer thin layer to the exact measurements of an AMD heatspreader and lay it inbetween my XP120 and the heatspreader. ftw!🙂 No seriously, I would love to see a silver based HS come out, but you know it would be $$. I thought I read somewhere that there was such a thing 😕
EDIT: Ahhh yes....Akaska came out with one...The Silver Mountain.
Click to expand...

Higher thermal resistance = bad. You want thermal conductance, which is the ability to conduct heat. You'd be better off cutting a disc out of rubber matting and using it, as it won't matter to your processor for very long that you used rubber instead of diamond.
Tas.
Click to expand...

Diamond is the best conductor. Why would this ^ (I know its far fetched but you get my point) or using diamond "dust" in a TIM not be the best conductor?
 
Originally posted by: JBDan
Originally posted by: BonzaiDuck
I happened across an archived project posted (I think) at OverClockers.com.

The poster had augmented his CPU HSF with a commemorative silver dollar, and posted some data on the the thermal resistance of substances in the Periodic Table of Elements.

He had lapped the coin's finish until it was smooth, bent the clips for his HSF slightly to accommodate the additional 1/8" thickness of the silver slug.

The lowest thermal resistance is provided by the following substances in ascending order (higher thermal resistance from top to bottom in the list🙂

Diamond
Gold
Silver
Copper
Aluminum

Thus the choice of silver in Arctic Silver thermal compound.

I was watching late-night TV the other night, and they were offering a commemorative 20-dollar gold-piece last minted during the Roosevelt Presidency. They were offering it for something like $37 through a 1-800 phone-number.

The Golden Eagle Tribute (?) coin is gold plated. It would seem that this would be all that's required to lower the thermal resistance of a heatpipe cooler like the XP-120 -- it might possibly be better than the recently released SI-120, and it would certainly improve the thermal resistance of the latter.

Comments or observations? Caller, You're on the Air . . . .
Click to expand...

I think I'll just grab my moms diamond bracelet, melt it down, make a wafer thin layer to the exact measurements of an AMD heatspreader and lay it inbetween my XP120 and the heatspreader. ftw!🙂 No seriously, I would love to see a silver based HS come out, but you know it would be $$. I thought I read somewhere that there was such a thing 😕
EDIT: Ahhh yes....Akaska came out with one...The Silver Mountain.
Click to expand...

Text🙂
 
That should make a helluva water-cooler.

We'd do well to search for further information on these metal properties.

Conductivity is measured in Watts per Meter-Kelvin (degree). Resistance is degrees Celsius or Kelvin per Watt of thermal energy.

That's an interesting point, and I'm not sure how it might make one material better than another, theoretically, but the silver is probably better because both properties are more conducive to efficient heat-transfer.

I think someone was correct about a coin with a core using a less desirable metal. I'd think it might depend on whether the plating of the coin extended around its edges, but that's just speculation.

And speculation with hopes of getting more insight through further discussion was my intention of posting this.

I don't think it would be hard to obtain a pure silver wafer, modify the ThermalRight XP/SI cooler of whatever model, and further reduce the idle-to-load temperature spread for any given processor and its maximum power or TDP specification. I also think that adding to the mass of the heatsink base with such a material would also reduce the idle temperature further, and the idle temperature would follow it, but I could be wrong here in this statement. Corrections and comments encouraged.
 
You'd be hard pressed to find something better than my favorite coin, the 1964 half dollar. It was the last year they made silver coins, and they put a lot of silver in it. Not to mention the 10% other material is copper. *eg* That would work VERY well for your experiment.

Not to mention they are rather plentiful, and you can find them on the cheap. Hell, you can get rolls of 20 for $50, and have some room to fvck it up. hehehehe
Tas.
 
Eventually, I may actually try that after I first bench-test the SI-120.

IF I damage the cooler, I could buy another one. I'm too curious to worry about the money at this point, although I do try and pinch pennies, but experiments have costs.

I bought a new replacement P4P800-SE motherboard the other day -- with a more contemporary BIOS than the original P4P800. I'll give the latter as a Xmas present with a Northwood and complete computer upgrade for my brother.

So I wanted to go ahead with testing the SE board with the Prescott 3.2E and a ThermalRight SI-120 cooler, with improvements to my motherboard duct. I had read that it's a good idea to "sink" your Mosfets, PLL chips, and Southbridge.

At that point, I wanted to use formidable heatsinks with tall spires, either through purchase or by cutting down a stock Intel P4 heatsink. I chose to buy the MicroCool sinks at Sidewinder, but discovered you could get less formidable ramsinks from SVC for $1.50 per package of ten. And then the issue arose about how to secure them and get better performance.

In the end, I accepted the risks of using Arctic Silver Thermal Adhesive, and I was lucky I coated my PLL pinouts with silicone before applying the Sink and ASTA, because my hand wasn't steady and the sink slipped, pushing ASTA over the edge of the PLL and onto (what I thought was) the solder pins. But I hadn't been confident enough in my own work. The silicone rubber completely protected them.

The questions are now three where there was once only one: (1) did other sloppy handling cause the ASTA to bridge two Mosfet pinouts? (Not likely, since they all got the silicon treatment on the electrical leads and solder pad.) (2) How much will my mobo temperature drop with the Mosfets, PLLs and Southbridge "sinked," and how much will my CPU idle temperature drop?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top