Thermal paste

Does using thermal paste on the CPU really make a great deal of difference
in conducting heat to be dissapated? What sort of temperature differences
are we talking about between using an average thermal paste and a CPU using
no paste whatsoever?

The difference between possibly burning out your CPU, and not...

--
DaveW



"-" wrote in message
...
Does using thermal paste on the CPU really make a great deal of difference
in conducting heat to be dissapated? What sort of temperature differences
are we talking about between using an average thermal paste and a CPU
using
no paste whatsoever?

Does using thermal paste on the CPU really make a great deal of difference
in conducting heat to be dissapated? What sort of temperature differences
are we talking about between using an average thermal paste and a CPU
using
no paste whatsoever?

Yes it does. Air does not conduct heat well at all, and that's what will be
sitting between you processor and heatsink unless you use thermal paste of
some sort. Here are the various thermal conductivities for some materials.
The smaller they are, the slower they transfer heat, and thus the hotter you
CPU will get. As you can see, air is like dressing you processor up in a
wooly jacket and mittens:

Air 0.02 W/mK
Conventional thermal paste ~ 1 W/mK
Silmore thermal paste ~ 2.80 W/mK
Arctic Silver II thermal paste ~ 8.00 W/mK
Arctic Silver III thermal paste ~ 9.00 W/mK
Aluminum 220.00 W/mK
Copper 382.00 W/mK

You'll also note that the thermal paste is considerably worse than the metal
of your heatsink for transferring heat, so you should be putting on as
little as possible: just enough to replace the air gaps between the two
surfaces.

Ewan

Ewan Sinclair wrote:

Does using thermal paste on the CPU really make a great deal of
difference in conducting heat to be dissapated? What sort of temperature
differences are we talking about between using an average thermal paste
and a CPU
using
no paste whatsoever?

Yes it does. Air does not conduct heat well at all, and that's what will
be sitting between you processor and heatsink unless you use thermal paste
of some sort. Here are the various thermal conductivities for some
materials. The smaller they are, the slower they transfer heat, and thus
the hotter you CPU will get. As you can see, air is like dressing you
processor up in a wooly jacket and mittens:

Air 0.02 W/mK
Conventional thermal paste ~ 1 W/mK
Silmore thermal paste ~ 2.80 W/mK
Arctic Silver II thermal paste ~ 8.00 W/mK
Arctic Silver III thermal paste ~ 9.00 W/mK


Where did you get these figures?? Several places I've seen on the net, they
found almost ZERO difference between "artic" thermal paste and the regular
white stuff sold at Radioshack for pennies.

http://thecrucible.ca/reviews/thermal/performance2.php

Actually in this test the cheapo radioshack grease outperforms ALL the Artic
silver products!! LOL!!

Another test I saw showed in a few days of use the artic stuff was already
breaking down and causing higher temps than regular type thermal compounds.

--

Stacey

"-" wrote in message
...
Does using thermal paste on the CPU really make a great deal of difference
in conducting heat to be dissapated? What sort of temperature differences
are we talking about between using an average thermal paste and a CPU
using
no paste whatsoever?



get a good magnifying glass
then have a look at your heatsink's surface

you may be surprised !

Where did you get these figures?? Several places I've seen on the net,
they
found almost ZERO difference between "artic" thermal paste and the regular
white stuff sold at Radioshack for pennies.

http://thecrucible.ca/reviews/thermal/performance2.php

Actually in this test the cheapo radioshack grease outperforms ALL the
Artic
silver products!! LOL!!

Another test I saw showed in a few days of use the artic stuff was already
breaking down and causing higher temps than regular type thermal
compounds.

Stacey

Hi.

That is most likely because they used to much of it, when you use Arctic or
similar cooler paste then you should only use a drop or two and then spred
it out in a very thin layer. Otherwise you will get the exact opposite
effect.

You don't get the same effect when using to much of normal cooler paste.

Regards
Martin Bech
Copenhagen, Denmark

On Tue, 11 Nov 2003 23:58:48 -0500, stacey wrote:


Air 0.02 W/mK
Conventional thermal paste ~ 1 W/mK
Silmore thermal paste ~ 2.80 W/mK
Arctic Silver II thermal paste ~ 8.00 W/mK
Arctic Silver III thermal paste ~ 9.00 W/mK


Where did you get these figures?? Several places I've seen on the net, they
found almost ZERO difference between "artic" thermal paste and the regular
white stuff sold at Radioshack for pennies.


The figures look about right, but relatively speaking, most of the
thermal compounds are close enough. Contrast them to air in a
worst-case or heatsink as best, and the thin, pressurized layer of
compound makes little difference so long as it's consistency allows it
to spread thin and even but not run out or degrade over time.


Dave

"-" wrote in message
...
Does using thermal paste on the CPU really make a great deal of difference
in conducting heat to be dissapated?

YES

What sort of temperature differences
are we talking about between using an average thermal paste and a CPU
using
no paste whatsoever?

TOO MUCH

Hank

Martin Nikolaj Bech wrote:

Where did you get these figures?? Several places I've seen on the net,
they
found almost ZERO difference between "artic" thermal paste and the
regular white stuff sold at Radioshack for pennies.

http://thecrucible.ca/reviews/thermal/performance2.php

Actually in this test the cheapo radioshack grease outperforms ALL the
Artic
silver products!! LOL!!

Another test I saw showed in a few days of use the artic stuff was
already breaking down and causing higher temps than regular type thermal
compounds.

Stacey

Hi.

That is most likely because they used to much of it, when you use Arctic
or similar cooler paste then you should only use a drop or two and then
spred it out in a very thin layer. Otherwise you will get the exact
opposite effect.

The only place I've seen Artic silver beat radio shack type paste is on
Artic silvers website or from people who paid $$$ for that stuff. The clear
REALLY cheapo paste isn't very good but any of the white silicon thermal
pastes work as well as the $$$$ ones do.



You don't get the same effect when using to much of normal cooler paste.


So normal cooler paste isn't a "problem" when it's too thick but the "high
conductive" stuff is?

--

Stacey

Where did you get these figures?? Several places I've seen on the net,
they
found almost ZERO difference between "artic" thermal paste and the regular
white stuff sold at Radioshack for pennies.

I just got them off a random Tom's Hardware Guide page. They throw them on
all the time. here's one:

http://www.tomshardware.com/cpu/20020916/cooler-02.html

I'd certainly say that they're a credible source.

Actually in this test the cheapo radioshack grease outperforms ALL the
Artic
silver products!! LOL!!

Aha, so this site would be a stealthy Radioshack marketing ploy then :-)

Another test I saw showed in a few days of use the artic stuff was already
breaking down and causing higher temps than regular type thermal
compounds.

I have heard of silicone based thermal pastes being subject to "creep" over
periods from 3 months to a year, where air gaps get left as the paste
shifts. Arctic Silver II was silicone based, but Arctic Silver III and
greater aren't. However, in the 3 day range, this is more likely to be the
compound adjusting before finally settling in. From the Arctic Silver 5
page:

"During the CPU's initial use, the compound thins out to enhance the filling
of the microscopic valleys and ensure the best physical contact between the
heatsink and the CPU core. Then the compound thickens slightly over the next
50 to 200 hours of use to its final consistency designed for long-term
stability."