Artic Silver Ceramique

DasFox

Diamond Member
Sep 4, 2003
4,668
46
91
For Artic Silver Ceramique compared to AS5 what is the density and thickness like? I'm not sure why, but I guess I was thinking it was a thinner feeling grease to the touch compared to how people where comparing it to AS5, but I see that this Ceramique is quite thick and sticky.

Just hope as everyone is saying it is alot easier to take the HSF off with this stuff.

THANKS
 

Fullmetal Chocobo

Moderator<br>Distributed Computing
Moderator
May 13, 2003
13,704
7
81
Originally posted by: DasFox
For Artic Silver Ceramique compared to AS5 what is the density and thickness like? I'm not sure why, but I guess I was thinking it was a thinner feeling grease to the touch compared to how people where comparing it to AS5, but I see that this Ceramique is quite thick and sticky.

Just hope as everyone is saying it is alot easier to take the HSF off with this stuff.

THANKS

I don't see how it is easier to take the CPUs off. I pull my Xeons out of the socket every time I pull the HSF off. But I've never done damage to them, so I don't mind too much. And I be as careful as I possibly can while doing it, so I don't try to twist it or anything. I just prefer Ceramique on large surface areas compared to AS5.
 

DasFox

Diamond Member
Sep 4, 2003
4,668
46
91
ok, yeah when I tried to pull one off last week with AS5 I couldn't get the thing out, man it was like pulling teeth. Is there any recommended way to pull the HSF off or get the cpu out with it in one piece?

They should make that CPU bar in another position for this, so that you can lift it up and pull it all out in one piece, easier.
 

keldog7

Senior member
Dec 1, 2005
235
0
0
not much experience taking the sink off afeter extended use (i.e. 1-2 years), but after a few weeks, its never been tough with the ceramique, like I've had with some of the "grey goo" types. (but then, who hasn't bent a few pins trying to force the heatsink loose)

as for application consistency...it reaaly varies with application temperature...
at room temp (20-22C), i'd compare it to corn syrup
but by rubbing your hand together, as if to warm them, and putting the ceramique tube in between your hands, you can very easily warm it (in 10-20 seconds) to a consistency like maple or Aunt Jem syrup. just remember that the chip will still be cool with this method, and it will get sticky as soon as applied
-A
ps. can anyone guess what I ate for breakfast?
 

DasFox

Diamond Member
Sep 4, 2003
4,668
46
91
Well this tube of Artic Silver Ceramique is a lot thicker then corn syrup, very thick and tacky feeling.

PANCAKES, LOL ;)
 

Edzard

Senior member
Jul 23, 2003
504
0
0
Minor edits, for clarity, February 15, 2006.

Here is my two cents worth.

I can't compare the "thickness" of Ceramique with AS5 because I've never used AS5. Compared to AS2, Ceramique is much "thicker", or perhaps more accurately, more viscous.

But before you use the stuff, read on...

I've used AS Ceramique on at at least six processors (four Socket A AMD Athlon XP CPUs and two S754 Semprons) and familiarity is most decidedly breeding contempt. On Socket A it is OK provided that you are willing to put up with its tendency to form "strings" as you pull the tube away from the top of the processor core after applying the compound. Those "strings" can be messy and flop over onto areas of the processor other than the core or heat spreader - hello cleanup (at least if you are as meticulous about this aspect as I am). Also it is not easy to spread the compound evenly over the heatsink which I believe is supposed to be done (the compound should be applied to both the core/heat spreader AND the part of the heatsink that will be contacted).

When you remove a Socket A (or Socket 370) processor that has been applied with Ceramique, the removal isn't very difficult because the Ceramique doesn't tend to glue the heatsink to the processor core due to the small areal contact between the core and the heatsink (for Socket 370, I'm assuming a Coppermine core which is roughly the same size as a Socket A processor core manufactured using the .13 micron process, but definitely not an earlier, larger S370 core or a Tualatin core with its heat spreader). So you can first remove the heatsink, lift the processor retention lever, and remove the processor -- normal procedure. And if the processor does stick to the heatsink, gentle rotation of the heatsink should suffice to break any core/heatsink bond and could probably be done while the processor was still in the socket (I've never had to do this when using Ceramique with Socket A processors).

BUT on heat spreader equipped processors that fit into a socket that is recessed due to the surrounding plastic bracket for retaining the heatsink (as in S754, S939, and many P4s), I absolutely do NOT recommend Ceramique. In these combinations (heat spreader CPU combined with 'recessed' socket) the much larger areal contact between the heat spreader and heatsink gives AS Ceramique the opportunity (which it seizes upon after "setting up" after two or so weeks) to start acting like a blooming bonding agent. Due to the lack of space between the heatsink and the surrounding bracket, it is not possible to rotate the heatsink back and forth horizontally in an attempt the break the Ceramique bond. So as mentioned in earlier posts, when you remove the heatsink, out comes the processor as well, stuck solidly to the base of the heatsink - definitely non-standard procedure. Then you have to rotate the processor back and forth in a semi-circular movement, holding the processor carefully at its edges, to get it to break free of the heatsink. Perhaps application of a solvent and/or heat would make this easier but you can't apply a solvent until the CPU and heatsink are out of the socket. After the processor breaks free of the heatsink, you have a shallowly grooved circle in the base of your heatsink (my experience with a retail AMD all-aluminum S754/S939 heatsink). And hopefully you haven't bent any pins or damaged anything else (like the socket or the retention mechanism) during the entire nerve-wracking process.

Perhaps removal could be facilitated by (1) first removing the retaining bracket, allowing one to rotate the heatsink to break the Ceramique bond (but what if the processor comes out of the socket during this process and pins get bent or you damage surrounding capacitors?); (2) overclocking the processor and running PRIME95 for 30 minutes or so in an attempt to heat up the heatsink and particulary the Ceramique so that it becomes more like syrup and less like bondo, then quickly attempt to remove the processor without it sticking to the heatsink or (3) use a heating device to warm up the heatsink and the Ceramique followed by hopefully the "clean/standard" removal of the processor. But why should it be necessary to bloody overclock your processor or use a blowtorch on your heatsink just to remove a processor - huh?

Have I missed something?

I have used, and liked, AS2. But thumbs down on Ceramique.


 

Edzard

Senior member
Jul 23, 2003
504
0
0
"Thickness" = viscosity in the context of the original post ("...what is the density and thickness like?"). There was no mention in my post of "density" and no equating of density to viscosity in either the original or my post. :D

And the traditional mathematical symbol for "not equal" is of course "?", although the symbol combinations of "!=" and "<>" have carried over from programming languages to widespread use in forums and newsgroups to denote the same meaning.....
 

keldog7

Senior member
Dec 1, 2005
235
0
0
Lets not get into a shouting match over semantics. I've only taken one material sciences course, and even I know enough to say that density is not viscosity. While I think I can define density accurately, I hesitate to try viscosity. Either way, Howard is right, and so is Edzard. The original poster asked about density, but clearly (DasFox, correct me if I'm wrong) wanted to know about the thickness of Ceramique...perhaps he / she should have asked for viscosity. Overall, it strikes me that the important parts of these posts is "how easy is it to work with Ceramique?...and how easy is it to remove the HSF after a bit of time passes. Good discussion. Apparently, I'm no longer looking forward to pulling my HSF off of my Opteron.
 

Witchfire

Senior member
Jan 13, 2006
226
1
0
Just a note on removing heatsinks from processors... The wife's hairdryer is my best friend when it's time for that job. Works like a charm every time, AS5, ceramique, Nanotherm PCM+, doesn't matter what I'm using.
 

Howard

Lifer
Oct 14, 1999
47,982
10
81
Originally posted by: Edzard
"Thickness" = viscosity in the context of the original post ("...what is the density and thickness like?"). There was no mention in my post of "density" and no equating of density to viscosity in either the original or my post. :D

And the traditional mathematical symbol for "not equal" is of course "?", although the symbol combinations of "!=" and "<>" have carried over from programming languages to widespread use in forums and newsgroups to denote the same meaning.....
Yeah, I was responding to OP.
 

Edzard

Senior member
Jul 23, 2003
504
0
0
Originally posted by: Witchfire
Just a note on removing heatsinks from processors... The wife's hairdryer is my best friend when it's time for that job. Works like a charm every time, AS5, ceramique, Nanotherm PCM+, doesn't matter what I'm using.

IMHO, an excellent contribution Witchfire, even though the discussion has veered a bit from "density and thickness" to the potential pitfalls of working with Ceramique and other compounds having the same "dense/gooey/thick/viscous" properties. But I suspect that DasFox, who started this thread, probably won't mind.

With regard to Ceramique that has been in place for several weeks or longer (on an installed processor) and thus had plenty of time to "set up", after heating with the hairdryer, does the Ceramique substantially return to its original gooey/syrupy state and thus enable truly easy removal of the heatsink, with no risk of the processor coming out with it? Please do give more details, especially with regard to how easy it becomes to remove the heatsink, sans processor, after heating via the hairdryer, and whether or not there are any special tricks (removing the fan before using the hairdryer comes to mind) or hazards (like overheating the processor) involved.

It isn't much of a leap to think that the hairdryer might come in handy when applying Ceramique and other gooey thermal pastes, especially to the heatsink, during installation of a processor (but applying heat directly to the processor could certainly be risky). Have you done that (used a hair dryer to get a more uniform application of thermal compound on a heatsink prior to installation) and would you please relate relevant details of your experience?

Thanks for your post.
 

Witchfire

Senior member
Jan 13, 2006
226
1
0
Edzard, I believe I'm at a bit of an advantage over most, in that the last three coolers I chose to use have allowed me to twist the heatsink as I remove it. The Ceramique seems to not turn back to it's original gooey state as much as turn more to the consistancy of a thick paste (think 1st grade posterboard paste) . The Thermalright XP120 I currently favor allows me to twist it off, and to apply the heat to the baseplate... As far as overheating the processor, all I can say is that I've never had a problem with a processor in any way after using the hairdryer to remove the heatsink. I did this process to my 3700+ 754-pin at least 6 times, and it never had a problem..

I've never used the hair dryer during application, so I'm not sure at all how that would work, but I agree that it sounds like a decent idea.

During application, I try to use as little TIM as possible, and have gone over to just putting a small dollop dead center of my processor & letting the heatsink flatten it out. I've watched for any evidence of an air pocket upon removal, but have seen none, and temps have always been fine. Using a hair dryer may well make the Ceramique easier to work with if you prefer to spread it out, but I've always subscribed to the 'simple is best' school of thought.. why complicate something that doesn't need to be?

Hope something here helps.