Sony demos silicon sheets that could replace thermal paste

Matthew DeCarlo

Posts: 5,271   +104
Staff

Although applying thermal grease isn't rocket science, there is some guesswork involved, especially if it's not something you do often. It's not uncommon to see questions about the application of thermal paste or even horror stories about the overuse of such compounds. For seasoned enthusiasts, the endeavor is less risky, but it's still a hassle to clean and reapply grease when tinkering with processors and coolers.

Tech-On reports that a new product shown last week in Tokyo by Sony could solve these issues. The company revealed a prototype of a silicon and carbon-based thermal sheet currently known as "EX20000C" that measures 0.3-2.0mm thick with a thermal resistance of 0.4-0.2K cm2/W and a compressive load of about 1-3kgf/cm2 -- five or six times the effectiveness of Sony's previous solution.

EX20000C supposedly meets, if not exceeds the thermal conductivity of standard thermal grease. A show floor exhibit compared the sheet's effectiveness with a conventional setup and it managed to reduce the processor's temperatures by an additional three degrees Celsius. In addition to eliminating cleanup and preventing overapplication, the sheets last longer than paste because they don't degrade.

The chips, coolers and grease involved in this demonstration weren't shared, or at the very least weren't reported by Tech-On. Sony expects the new sheet to be used in high-end servers, projectors and other such machines and it's unclear when or if the solution will be available for your next system build and how much they might cost.

Permalink to story.

 
this is very interesting, only the so called average conventional greese can range from 0.5 degree of difference to 6 degrees. it'll be 'really' helpful to know what kind of conventional greese was used for the test.
 
Not to mention high end compounds can out-pace the average Joe grease by at least 3 degrees as it is. My guess is they're comparing it to their own "best case scenario" paste, which chances are isn't all that good. I'd like to see it compared to IC Diamond or Indigo Extreme to get a real comparison.
 
You know, even if it isn't as good as the best thermal compounds and even if it isn't sold to the average consumer, we can ALL feel more comfortable with the fact that OEM's can consistently apply the same amount of thermal grease to each and ever processor they use or produce. May I remind everyone of Apple's literally botched grease efforts in their older Macbook Pros?
 
@ spydercanopus: those are made of thermal grease. The "new" part is instead of using thermal grease, they use [FONT=Helvetica]silicon and carbon-based thermal sheets. [/FONT]
 
This comment will probably generate some responses telling me I don't know what I'm doing. Maybe I don't....

I've used Arctic Silver 5 and Arctic Silver Ceramique on maybe 10 different occasions. I've also used Radio Shack's "thermal compound". Radio Shack's stuff is much easier to apply and I haven't noticed any 'bad' temps by using the RS stuff. I don't overclock, so perhaps that is some of it. But I'm not really convinced the expensive stuff is worth it.

Maybe I just apply wrong, but I follow the instructions given and I've watched some youtubes and whatnot.
 
This comment will probably generate some responses telling me I don't know what I'm doing. Maybe I don't....

I've used Arctic Silver 5 and Arctic Silver Ceramique on maybe 10 different occasions. I've also used Radio Shack's "thermal compound". Radio Shack's stuff is much easier to apply and I haven't noticed any 'bad' temps by using the RS stuff. I don't overclock, so perhaps that is some of it. But I'm not really convinced the expensive stuff is worth it.

Maybe I just apply wrong, but I follow the instructions given and I've watched some youtubes and whatnot.
At stock frequencies you really won't see that much of a change. Only a few degrees C. Its far more important that you actually use some type of heat paste and that you apply it correctly without air bubbles.
 
This comment will probably generate some responses telling me I don't know what I'm doing. Maybe I don't....

I've used Arctic Silver 5 and Arctic Silver Ceramique on maybe 10 different occasions. I've also used Radio Shack's "thermal compound". Radio Shack's stuff is much easier to apply and I haven't noticed any 'bad' temps by using the RS stuff. I don't overclock, so perhaps that is some of it. But I'm not really convinced the expensive stuff is worth it.

Maybe I just apply wrong, but I follow the instructions given and I've watched some youtubes and whatnot.
At stock frequencies you really won't see that much of a change. Only a few degrees C. Its far more important that you actually use some type of heat paste and that you apply it correctly without air bubbles.

And perhaps being able to apply this new Sony silicon sheet consistently is the reason I am excited about this news, besides the fact that there is no clean-up and having to guess how much grease to use depending on the contact surface area of the CPU and heat sink. I just hope that Sony will make it available to the consumer market as well.
 
People have mentioned they struggle to lay new paste etc but there was nothing to stop you buying the pre measures sheets applications that Akasa made if you failed so hard at the old cardboard routine.
 
I would think if you used cardboard to spread the paste you were already failing. You have a pretty high risk of flaking fibers from the cardboard off into the paste. Credit card or other similar flat edged plastic would be a better choice.
 
[FONT=Helvetica]The last time I measured pressure, I had a result in pascals (Pa) - same as N/m^2 (I'm an engineer). So what the hell does 1-3kgf/cm2 mean? Is that the same as the bending strength of an elephant's trunk?[/FONT]
[FONT=Helvetica]Come on... give it to us in actual, recognised units. I can understand 'Pa', 'bar' or 'N/cm^2' but "kgf" is totally new to me.[/FONT]
[FONT=Helvetica]And no, kilogram-force is no more a unit than 'cubic pounds per arc-minute' is a unit. Just sayin'.[/FONT]
 
Really cool SONY. Lets hope this doesnt backfire in the PS4 so please do all the necessary test
 
Unless it comes at less than or at the same price, and provides more than 5C of temp drops, it's not worth it.
I would think if you used cardboard to spread the paste you were already failing. You have a pretty high risk of flaking fibers from the cardboard off into the paste. Credit card or other similar flat edged plastic would be a better choice.
Both the methods are inefficient/dangerous, due to the uncertainty in the final thickness and cover area of the paste and inevitable air-bubbles. I've covered that in EACG-II. You can check it out if you like from the link in my sig.
I haven't replaced my paste in a while, but when I do, I'll attach some pics to show ideal application. Covering the Don'ts would be rather expensive for me.

I for one use generic paste, and with proper application alone I shaved off 4C. Nothing high end.
One should note that the temp-freq graphs aren't linear. Once you exceed the stock speeds, the scene pretty much changes drastically. The worst-case scenario is in which one should (I do) look for differences, and not on a happy system.
 
I deed some of that for my head for this summer for my brain needs to run 3 degrees cooler....
 
Back