Thermal Compound Spreading

On Sun, 18 May 2008 11:29:41 -0500, "Ed Medlin" <ed@ edmedlin.com>
wrote:

>
>"Phil Weldon" <notdisclosed@example.com> wrote in message
>news:sIGdnbX6_tcv87LVnZ2dnUVZ_tXinZ2d@earthlink.c om...
>> 'John Whitworth' wrote:
>>> What do you use instead? Regular compound?
>> _____
>>
>> Whatever is around at the time. I've been working off a 2 ounce tube of a
>> zinc oxide filled polyester base thermal compound (about the consistency
>> of Vaseline) for years. I keep threatening to get some 100,000 mesh
>> diamond polishing powder and make my own thermal compound - that's about
>> the only non-poisonous, non-metallic filler that is significantly better
>> than zinc oxide (actually more heat conductive than any metal, and
>> surprisingly inexpensive). 100,000 mesh powder is about the feature size
>> of the Pentium 60 B^)
>>
>> Honestly, when I did run a set of comparisons back in the days of the
>> Celeron 300a (the first CPU that easily overclocked by 50 %), I posted the
>> results in alt.comp.hardware.overclocking. I used the original Arctic
>> Silver, RadioShack generic zinc oxide filled silicone grease, no compound
>> at all, and unsalted butter. Except for the 'no compound at all', only a
>> couple of degrees C separated the performance of the various compounds.
>>
>> Phil Weldon
>>
>I remember those tests......:-). Someone here actually used a thin coating
>of Skippy Creamy Peanut Butter and got the same results, although he said it
>got a bit "rank" smelling after awhile...:-). I use whatever compound I have
>close by. My local 'puter shop gives me Arctic Silver whenever I get
>anything there, so I have a lot of it around and plenty of Radio Shack zinc
>oxide and I use whatever I grab first. I honestly can't see any difference
>unless I accidently apply too much Arctic Silver and then I will see
>slightly higher temps. I don't get that with the zinc oxide...... It just
>gets messy......:-).

As misfit pointed out there may be an issue with using other compounds
such as butter or peanut butter. They may do the same cooling as the
specially made compounds when initially applied, but how well will it
work a year later? If the compound dries out it's unlikely to be
working well as a thermal conductor. So I think I will stay away from
putting peanut butter on my CPUs and stick with a compound that's made
for the purpose (since it supposedly won't dry out over time.)

These natural products are also subject to air oxidation (rancidity)
which produces acidic decomposition products.

AS5 has always worked flawlesly for me, but it's terrible with jelly
on a sandwich!!!

On Wed, 21 May 2008 13:51:40 -0400, shawn <nanoflower@gmail.com>
wrote:

>On Sun, 18 May 2008 11:29:41 -0500, "Ed Medlin" <ed@ edmedlin.com>
>wrote:
>
>>
>>"Phil Weldon" <notdisclosed@example.com> wrote in message
>>news:sIGdnbX6_tcv87LVnZ2dnUVZ_tXinZ2d@earthlink. com...
>>> 'John Whitworth' wrote:
>>>> What do you use instead? Regular compound?
>>> _____
>>>
>>> Whatever is around at the time. I've been working off a 2 ounce tube of a
>>> zinc oxide filled polyester base thermal compound (about the consistency
>>> of Vaseline) for years. I keep threatening to get some 100,000 mesh
>>> diamond polishing powder and make my own thermal compound - that's about
>>> the only non-poisonous, non-metallic filler that is significantly better
>>> than zinc oxide (actually more heat conductive than any metal, and
>>> surprisingly inexpensive). 100,000 mesh powder is about the feature size
>>> of the Pentium 60 B^)
>>>
>>> Honestly, when I did run a set of comparisons back in the days of the
>>> Celeron 300a (the first CPU that easily overclocked by 50 %), I posted the
>>> results in alt.comp.hardware.overclocking. I used the original Arctic
>>> Silver, RadioShack generic zinc oxide filled silicone grease, no compound
>>> at all, and unsalted butter. Except for the 'no compound at all', only a
>>> couple of degrees C separated the performance of the various compounds.
>>>
>>> Phil Weldon
>>>
>>I remember those tests......:-). Someone here actually used a thin coating
>>of Skippy Creamy Peanut Butter and got the same results, although he said it
>>got a bit "rank" smelling after awhile...:-). I use whatever compound I have
>>close by. My local 'puter shop gives me Arctic Silver whenever I get
>>anything there, so I have a lot of it around and plenty of Radio Shack zinc
>>oxide and I use whatever I grab first. I honestly can't see any difference
>>unless I accidently apply too much Arctic Silver and then I will see
>>slightly higher temps. I don't get that with the zinc oxide...... It just
>>gets messy......:-).
>
>As misfit pointed out there may be an issue with using other compounds
>such as butter or peanut butter. They may do the same cooling as the
>specially made compounds when initially applied, but how well will it
>work a year later? If the compound dries out it's unlikely to be
>working well as a thermal conductor. So I think I will stay away from
>putting peanut butter on my CPUs and stick with a compound that's made
>for the purpose (since it supposedly won't dry out over time.)

"John Whitworth" <sexyjw@g_EEEEEEEEEEEEEEEEE_mail.com> wrote in message
news:482f246f$0$26085$db0fefd9@news.zen.co.uk...
> Just found the following article. What's everyone's views?
>
> http://www.overclockers.com/articles1529/
>
> To date, I have been a religious credit card spreader!
>
> JW

I just put a dab of the radio shack stuff in the middle, about the size of a
booger. I always see a square of white on the HS when the CPU is removed so
I know it works.

On Wed, 21 May 2008 17:26:44 -0500, Al Brumski <A_Brumski@least.ok?> wrote:
: These natural products are also subject to air oxidation (rancidity)
: which produces acidic decomposition products.

Unsalted butter has the advantage of being usable with an uninstalled
processor affixed to the topside of a cat. You never know when you'll
need ap erpetual motion machine