Arctic Silver Vs. Ultra Silver???

[fabioschutz]

What is the BEST thermal grease: Arctic Silver or Ultra Silver ????

Anyway anyone know something or have some experience about the Ultra Silver????

Tks...

 

[mikee]

Where did you see this Ultra Silver stuff?

 

[fabioschutz]

Selling in ColdCPU.
And its more expensive than Arctic Silver, and they have said a lot about it. But I really dont know...

See here: http://www.coldcpu.com/product.asp?alphacatid=THERMGR

 

[Sunny129]

i've never heard of ultra silver until now. but to be honest with you, arctic silver is about the best stuff out there, so if this ultra silver stuff is better, it cant make that significant a difference since they are both based on silver...with a difference of $1, it really doesnt matter what you get, unless you want to go with arctic silver which is known to be great stuff...

 

[bonkers325]

ultra silver lol

cheers,
bonkers325

 

[jimmygates]

Anyone have any idea on who manufacturers this "Ultra Silver"? There are no specs posted on that site. I'm curious to see results if testers compare this to Artic Silver.



-Jimbo

 

[Mikewarrior2]

I asked coldcpu for the MSDS(chemical breakdown document) for Ultra Silver, or for a link to the manufacturer so I can ask them. So far, no response(and no, it isn't wrong ot ask for this document, it is available for pretty much anything).


Mike

 

[DaddyG]

I wouldn't touch Ultra Silver with a 10 foot pole. We know the efforts that Nevin has gone through with AS. BTW just because something is silver means nothing. Keeping the silver in suspension, coating the silver to make sure its not conductive, making the compound stable to 400f (if i'm not mistaken). All this contributes to a good product.

 

[bonkers325]

right on

cheers,
bonkers325

 

[buildingacomputer]

My apology for ignorance first.

Does Arctic Silver truly have silver flakes? How can a paste containing silver particles be electrically non-conductive? The only way to do this (electrically non-conductive) is to limit the number of silver particles so that they do not touch one another. If particles are not touching, it cannot be an excellent thermal conductor. If particles are not touching, why not use aluminum flakes? What about copper flakes? Why silver? To me, this is a mystery. I almost feel that someone is cheating us. Does anyone have a website where this is explained?

 

[Mikewarrior2]

Artic silver is not "silver-flakes" based, rather it is micronized silver particles. it is 80-83% micronized silver, with 13-16.5% silicone oil. It isn't conductive(electrically speaking), because the silver is kept "isolated" to prevent electrical conduction, except under situations where it is under extreme compression.

Hopefully nevin can come here and explain this further, but it is not electrically conductive in its normal state, nor in its applied state.

Also, silver is much better than copper or aluminum for thermal transfer.


Mike

 

[olman]

Actually the specs for thermal conductivivity of Ag and Cu are as follows plus some other metals and arctic silver: The list goes from greatest to least.

Ag 429 Wm^-1K^-1*3
Cu 401 Wm^-1K^-1*3
Au 317 Wm^-1K^-1*3
Al 237 Wm^-1K^-1*3
Arctic Silver >8 Wm^-1K^-1*3

So as you can see Ag isn't really that much more thermally conductive than Cu. Now if you were to say Ag is much more conductive than Arctic Silver then that would be more realistic.

I am checking out a possible combination of a liquid metal(not mercury), Ag powder, and silicone base as we speak. I made it today and it is on my ocd ppro, you know how big that chip is. Lots of surface area. I would not try and use mercury as it evaporates at a low temp and can cause a brain disorder This metal evaporates at 1300+ deg C. Mercury also has a very low thermal conductivity measurment around arctic silver.

The liquid metal I am testing is a alloy and stays as a liquid to -20 deg C I guess. This liquid metal also can dissolve silver and I believe leave it as a solution with out hardening. The average ThermCond of this metal is 50 Wm^-1K^-1*3 which as you can see is way above arctic silver. With Ag added to the alloy it could well approach Al for therm-cond. I am sure arctic silver is very good I have just been making my own and want something even better. As buildingacomputer said wouldn't it be better if the metal particles were touching one another? Well yes it would and that is what I am attempting with this venture. My computer has been running for awhile now and at low temps so I am sure that it will work good. I didn't make this paste as good as it could have been for lack of metal but I will acquire a good lot soon and check different amounts to see what the best combo is. Also you wouldn't for sure want this to touch your circuits in any way as it would have high electrical conductivity.

ol' man

 

[Mikewarrior2]

olman,

See, the problem is, copper greases are no more than 25% copper antisieze compound, and same with most aluminum greases.

out of the existing thermal greases, circuitworks silver grease and arctic silver are pretty comparable, with arctic silver easily getting the nod due to its lack of electrical conductivity.

Of course, under ideal conditions, a perfect cpu to copper/aluminum connection would be far better than grease, but that would require lapping many many more times accurate than can be done by hand.


Mike

 

[olman]

May I ask why a person needs non-conducting thermal grease if they are really careful when they put it on. I am trying to get past having to lap the crap outa your HS/WB and CPU to get a real good thermal conducting joint

ol' man

 

[Mikewarrior2]

Duron "bridges"... usually, conductive grease isn't a problem. Some people, no matter how careful, splash grease around.

In general, its just a safety issue. In My experience, arctic silver performs slightly better than circuitworks silver, so I stick with AS. No worries in case i do use a little too much.


Mike

P.S. Its also useful if you to apply grease to your new heatsink on your video card

 

[olman]

If one was real worried about the bridges couldn't one cover them with tape until they are done or even for good. I have made Cu paste and it worked real good. I simply added Cu powder to a silicone paste. I would say it was around 50% to 60% Cu by weight. That brings up a question I have. When they say 80% to 84% silver, do they mean weight or volume. This would definatly be nice to know.

ol' man

 

[Mikewarrior2]

80-84% micronized silver by Weight.



Mike

 

[DaddyG]

Olman,

You also get into the quality of the silicone paste. Silicone is known to dry out over time, the heat and clip pressure of Socket A shortens the time dramitically.

 

[olman]

If the Thermal paste was made completely out of metal then it couldn't dry out. The BP. of the substance I am talking about is at about 2300 deg C. That is very high indeed only a few degrees below silver even.

ol' man