It's not rocket science although there is a bit of a learning curve.
Done properly, there is no excuse for air bubbles.
Put a pile of heat sink compound in the center of the metallic cover on
the CPU. Apply the heat sink straight down onto the compound with a bit
of a back and fourth twisting motion. You may have to do this several
times to get the proper amount. When done correctly the two surfaces
will hold on to each other with out excess flowing over the side.
Be willing to sacrifice some compound if you've not done this before.
Note, too thick a layer will greatly reduce the compounds ability to
transfer heat!
Cost? The best only costs a few cents more per application than the
cheap stuff and is substantially cheaper than the old Silicone thermal
grease of the early days. I had a 10 oz tube of that stuff (I worked for
a company that made it) that cost me a buck. I still have most of it,
but the tube is getting pretty beat up. I haven't used any of that
since we quit using those little "screaming" 90 mm fans. They were
worse than the air noise from a 10 KW amp.
BTW, I actually did find "Silicone grease" offered. It was an
unbelievable $4.99 for a 1 gram pack. 11 or more dropped it to $4.45. A
2.7g applicator of Arctic Silver Ceramique sells for $4.99 A 20 gram
applicator is $8.99. The same amount silicone grease would be $89. Get
any of these in bulk and you have handling and application issues.
Price depends highly on packaging. Having worked in the industry, as a
SWAG I's say the packaging cost several time the cost of the silicone
grease inside.
Go to www.newegg.com and search on "Heat transfer compounds". It's by no
means the comprehensive list of what they carry, but give an idea as to
what's available and costs. You can find a lot more with detailed searches.
Viscosity? Forget it and don't worry about it unless you make your ow,
or purchase large lots. Virtually ALL commercial compounds are of the
correct viscosity that let you get a "uniform" layer measured in
microns, "without air bubbles. The mor common compounds do cover a
variety of viscosity as do OEM bulk shipments, meaning the average user
has a lot more knowledge of metal to metal and metal to ceramic
interfaces that the typical ham or computer enthusiast.
How many know the difference between 10 Cs and 6000 Cs?. 10Cs is for
dry fly treatment while 6000 is like Taffy. What's 6000 used for. I
saw it in the labs but I retired in 97 and have no recollection as to
its use.
Why use the good stuff for less demanding applications when there is
little, if any difference in cost? I'd say, ease of installation with
the knowledge you used good stuff. When using the "good stuff" you
simply put "enough" on the surfaces and stick the together, knowing they
will work. Less efficient compounds may require more care when used.
The less efficient is likely to become difficult to find as the primary
use dries up. However, when it comes to price, it is to their advantage
to use the "cheapest" that "should" do the job, often relying on
manufacturers claims as to the product's characteristics.
Industry generally abandons a product before it reaches the point of
negative return.
BTW, if you lap a pair of steel (or any other metal surfaces) optically
flat and put them together (dry), they will be extremely difficult to
get apart. They stick so well that you may not be able to align the
edges. Left in this state, some metals will weld themselves together.
These compounds allow for small irregularities to exist in that metal to
metal interface and still transfer in areas that would have not been in
intimate contact.
Getting heat sink and CPU surfaces optically flat is labor intensive and
impractical. Keeping them in that state through time, handling, and
temperature changes would require special handling as well as training
of anyone who handled them, making the cost much higher than today's
mass produced CPUs, GPUs, and cooling devices.
73
Roger (K8RI)
On 4/11/2015 1:54 PM, Mike Waters wrote:
This is generally good advice. However, as Manfred nicely pointed out,
"...applying an ultra thin, even layer is almost a guarantee for entrapping
air bubbles. And that's definitely worse than a thick grease layer!".
I fully agree with Manfred. It depends on the viscosity of the grease,
whether it's thin enough to be easily squeezed out by applying pressure.
73, Mike
www.w0btu.com
On Sat, Apr 11, 2015 at 9:21 AM, Kevin Stover <kevin.stover@mediacombb.net>
wrote:
If after securing the transistor to the heat sink or heat sink to a
processor you see heatsink compound oozing out the sides you've used waaay
to much.
The layer of heatsink compound needs to be an even layer almost thin
enough to see through, with NO air bubbles.
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
---
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|