How much silicon paste?

  • Thread starter Thread starter Pete
  • Start date Start date
I was tempted to suggest that he coat the entire motherboard with paste, as a way of
insulating it against accidental short circuits caused by dropping his screwdriver
while working in the case with the computer plugged in and running ... but there's
always a chance that I'd be taken seriously.

Actually a fair number of board returns are due to surface mount
capacitors being chipped off the PCB, and other track damage.

I recall the Molex Skt 370 coolers which were a case of russian
roulette between destroying the table, your eye, the m/b or all.

Spray cream is not heatsink paste, it looks similar, but isn't.
 
Pete said:
AMD Athlon-64 3400
The CPU has a large metal area on top of it and doesn't
have the usual small square in the middle where you
concentrate the thermal paste.

So how should I put the paste on this CPU? How much, what
sort of spread and where should I concentrate it?

Ideally just slightly more than needed to completely fill the gaps
between the CPU package and heatsink bottom, and to see if you're
applying enough, spread a thin layer all over the CPU's metal top,
press the heatsink straight down without clamping it, then lift it
straight up. Any empty areas on the heatsink mean you didn't apply
enough. But too much paste, unless you really pour it on, is far, far
more desirable than even slightly too little since any excess will
squeeze out when you press on the heatsink while swirling it around
slightly. But if the paste is unusually viscous, warming the heatsink
with a hair dryer will help it flow out.
 
Dorothy said:
Plain heatsink compound will do fine, silicone free can
be useful as silicone technically can affect electronic
components if it comes into direct contact with them.
Obviously it's moot in this application unless your mode
of application is a large pump-action water pistol.

As far as I know, silicone oil is beneign to electronic components,
which is why heatsink compounds were always made with it until
recently. The move away from silicone oil to ester oil (what
silicone-free compounds use) was done mainly to prevent silicone
contamination of the equipment used for washing circuit boards after
assembly.

Here's a database of chemical compatibilities for various materials:
www.coleparmer.com/techinfo/chemcomp.asp
 
Dorothy Bradbury wrote:




As far as I know, silicone oil is beneign to electronic components,
which is why heatsink compounds were always made with it until
recently. The move away from silicone oil to ester oil (what
silicone-free compounds use) was done mainly to prevent silicone
contamination of the equipment used for washing circuit boards after
assembly.

Here's a database of chemical compatibilities for various materials:
www.coleparmer.com/techinfo/chemcomp.asp

Y'all just remember I did not, and do not, recommend any particular
compound. I merely pointed the originator to some instructions with a
recommend method of applying the crap. I don't care if he uses
toothpaste, shaving cream, KY jelly, whipping cream, marmalade, lemon
curd, or whatever!!!
 
Dorothy Bradbury said:
Actually a fair number of board returns are due to surface mount
capacitors being chipped off the PCB, and other track damage.

I recall the Molex Skt 370 coolers which were a case of russian
roulette between destroying the table, your eye, the m/b or all.

Spray cream is not heatsink paste, it looks similar, but isn't.

I read that as, "Spray cheese".

I wonder what the thermal characteristics of spray cheese are?

Jon
 
Actually a fair number of board returns are due to surface mount
I read that as, "Spray cheese".

I wonder what the thermal characteristics of spray cheese are?

Jon

Probably fair, but it would dry out. That's the problem with
everyday household substances. Now that I think of it, spray
cheeze would be full of tiny air bubbles, so maybe not so good.
 
Rubbish spouted by people selling "very special mega performance
Antartic mega god paste (tm)"

Hell, even toothpaste performs a similar role..

The fans are the same ones that will pay $200 a foot for audio
interconnect cables.

:-)
 
Y'all just remember I did not, and do not, recommend any particular
compound. I merely pointed the originator to some instructions with a
recommend method of applying the crap. I don't care if he uses
toothpaste, shaving cream, KY jelly, whipping cream, marmalade, lemon
curd, or whatever!!!

I use lox vomit.
 
On Wed, 26 Jan 2005 19:45:38 GMT, "Dorothy Bradbury"

o Thermal compound is a better heat transfer media than air
o So you simply need enough paste to fill the microscopic voids
Plain heatsink compound will do fine, silicone free can be useful
as silicone technically can affect electronic components if it comes
into direct contact with them.
<snip>

For what is worth, but in case someone finds it intersting. A while
back, in a different group, a fellow that seems to know quite a bit
recommended "a copper based anti-seize compound". He said:
"you can buy a copper based anti-seize compound at any auto parts
store; and for $5-6 it will do more cpu's than you will see in your
lifetime!"

Geo
 
Black Shuck said:
Hell, even toothpaste performs a similar role..

http://www.dansdata.com/goop.htm

Is that the review where the guy didn't give the Arctic its 200 hour break
in, with off-on cycles?

Other reviews have found that Arctic works, especially Ceramique.


--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.
 
GEO" [email protected] said:
On Wed, 26 Jan 2005 19:45:38 GMT, "Dorothy Bradbury"





<snip>

For what is worth, but in case someone finds it intersting. A while
back, in a different group, a fellow that seems to know quite a bit
recommended "a copper based anti-seize compound". He said:
"you can buy a copper based anti-seize compound at any auto parts
store; and for $5-6 it will do more cpu's than you will see in your
lifetime!"

Geo
nickel based works good also, I've used it many times. The metal
particles aren't as finely ground as they are in HS paste but what the
hell, it still works. I just traded in a t-bird 1000mhz that i put
together 4 years ago, blew the dust off the HS and tested it. Was at the
same temp reading as it was when I built it. It's funny how so many
people get so worked up about a little gob of goo
 
On Wed, 26 Jan 2005 19:45:38 GMT, "Dorothy Bradbury"



<snip>

For what is worth, but in case someone finds it intersting. A while
back, in a different group, a fellow that seems to know quite a bit
recommended "a copper based anti-seize compound". He said:
"you can buy a copper based anti-seize compound at any auto parts
store; and for $5-6 it will do more cpu's than you will see in your
lifetime!"

Geo


Copperkote
 
Randy Howard got up from the bar and shouted: :
The fans are the same ones that will pay $200 a foot for audio
interconnect cables.

:-)

Indeed, with gold plated connectors, and oxygen free cable, that when it
gets inside the boxes, runs along copper tracks, and up tin/lead solder
joints...
 
Is that the review where the guy didn't give the Arctic its 200 hour break
in, with off-on cycles?
My machine has maybe 2-3 off-on cycles a year. And most of them are for
long power failures. I've got a UPS, but not a generator or it would be
less than that.
Other reviews have found that Arctic works, especially Ceramique.

The all work, just none really that much better than the other.
 
nickel based works good also, I've used it many times. The metal
particles aren't as finely ground as they are in HS paste but what the
hell, it still works. ...

Now I'll have to compare prices. I hadn't heard of nickel based.

Geo
 
Take a look at the AMD website, I believe they have some
instructions/advice. With these new chips I believe that you don't put
the compound directly on the die, you put it on a casing around the
die. I think you'd want just a thin layer covering the whole of the
casing.
 
Black said:
Randy Howard got up from the bar and shouted: :


Indeed, with gold plated connectors, and oxygen free cable, that when
it gets inside the boxes, runs along copper tracks, and up tin/lead
solder joints...

You don't have your amps hand-built with gold connectors ? Tut, tut.
 
Black Shuck said:
Randy Howard got up from the bar and shouted: :


Indeed, with gold plated connectors, and oxygen free cable, that when it
gets inside the boxes, runs along copper tracks, and up tin/lead solder
joints...

Of course the manufacturers of high quality and expensive audio equipment
often use substandard components inside the box.

According to you?
 
Back
Top