Arctic Silver 5 or Zalman White Thermal Grease

[NEM]

Hi,

I'm in the process of replacing the stock Intel HSF with the Zalman
CNPS7000b all copper HSF. The processor is a P4 3.0c (overclocked at
3.3GHz) and is sitting on an ASUS P4C800-E Deluxe v2 motherboard.

What I'm wondering is which would be better to use; the included Zalman
White Thermal Grease or a one year old (sealed tightly-seems fine)
Arctic Silver 5? Can Arctic Silver 5 really go bad if kept in a zip bag?

Additionally, the BB size amount used (according to the Arctic Silver
instruction) in the center of the CPU doesn't mention pre-spreading
across the top of the CPU as the Zalman instruction indicate. Whichever
paste is used, would it be better to let the HSF spread the BB size
amount when clamped down or would it be better to pre-spread before
applying the HSF?

 

[Chuck]

Hi,

I'm in the process of replacing the stock Intel HSF with the Zalman
CNPS7000b all copper HSF. The processor is a P4 3.0c (overclocked at
3.3GHz) and is sitting on an ASUS P4C800-E Deluxe v2 motherboard.

What I'm wondering is which would be better to use; the included Zalman
White Thermal Grease or a one year old (sealed tightly-seems fine)
Arctic Silver 5? Can Arctic Silver 5 really go bad if kept in a zip bag?

Additionally, the BB size amount used (according to the Arctic Silver
instruction) in the center of the CPU doesn't mention pre-spreading
across the top of the CPU as the Zalman instruction indicate. Whichever
paste is used, would it be better to let the HSF spread the BB size
amount when clamped down or would it be better to pre-spread before
applying the HSF?

The Artic Silver 5 is far superior to any generic or even name brand thermal
paste or silver compound. For eample I swapped out the generic silver
compound you can buy at any CompUsa over the counter with Artic Silver 5 for
an athlon XP 2000+ palomino.. it ran at 43C at idle with AS5 compared to 55C
at idle with the generic stuff. Or as another example it is keeping my
Athlon XP 3200+ Barton at a nice cool 31C and that's with a case temp of 30C
and a room temp of 23C (74F). as for it going bad.. never heard that. as for
the amount to put on and application put a BB sized drop on th eHeatsink
itself then put 1/4 of that on your CPU and spread it to an extremely thin
layer spread the remaining amount and spread it across the heatsink, about
as thick as plastic wrap, which BTW is an excellent thing to use to spread
it.

 

[Larry Gagnon]

Hi,

I'm in the process of replacing the stock Intel HSF with the Zalman
CNPS7000b all copper HSF. The processor is a P4 3.0c (overclocked at
3.3GHz) and is sitting on an ASUS P4C800-E Deluxe v2 motherboard.

What I'm wondering is which would be better to use; the included Zalman
White Thermal Grease or a one year old (sealed tightly-seems fine)
Arctic Silver 5? Can Arctic Silver 5 really go bad if kept in a zip bag?

Additionally, the BB size amount used (according to the Arctic Silver
instruction) in the center of the CPU doesn't mention pre-spreading
across the top of the CPU as the Zalman instruction indicate. Whichever
paste is used, would it be better to let the HSF spread the BB size
amount when clamped down or would it be better to pre-spread before
applying the HSF?

I wouldn't get too anal about it. Studies show that expensive thermal
greases are rarely an improvement on regular inexpensive thermal heat sink
compound which you can get at any electronics store for next to nothing.
We have been thoroughly marketed to! At best you might get a 2 deg C
improvement. Big deal!

I would pre-spread the paste. Then you know for certain its thin layer
completely covers the die.

Larry Gagnon, A+ certified tech.

 

[DaveW]

The Arctic Silver is far better than white thermal grease, and, yes, spread
it around.

 

[JTS]

The Artic Silver 5 is far superior to any generic or even name brand
thermal paste or silver compound.

How do you know? Have you tried them all?

 

[Chuck]

How do you know? Have you tried them all?

It's impossible to try all thermal pastes, However I do know the results of
a 12C difference on exactly the same Processor and heatsink between standard
retail silver paste and the Artic silver 5 makes a huge statement for how
superior it is. especially since the 55C reading was just before shutting
down the system, then puling the heatsink and processor, cleaning both with
rubbing alchohol, applying AS5 and reinstalling the same processor and
heatsink, then powering it back up and letting it idle for 1/2 hour for a
temperature reading.. with a 43C reading a 12C drop from only changing the
thermal paste.

 

[NEM]

Hi Chuck,

The Artic Silver 5 is far superior to any generic or even name brand
thermal paste or silver compound. For eample I swapped out the generic
silver compound you can buy at any CompUsa over the counter with Artic
Silver 5 for an athlon XP 2000+ palomino.. it ran at 43C at idle with
AS5 compared to 55C at idle with the generic stuff. Or as another
example it is keeping my Athlon XP 3200+ Barton at a nice cool 31C and
that's with a case temp of 30C and a room temp of 23C (74F).

That's interesting that it dropped so much. I'm seeing nearly 60c under
a full load. Perhaps the Zalman 7000b will be the greatest benefit over
the stock Intel HSF, followed closely with the AS5. I'm hoping for
cooler and quieter operation.

as for it going bad.. never heard that.

as for the amount to put on and application put a BB sized drop on th
eHeatsink itself then put 1/4 of that on your CPU and spread it to an
extremely thin layer spread the remaining amount and spread it across
the heatsink, about as thick as plastic wrap, which BTW is an
excellent thing to use to spread it.

Thanks for the tip. Any suggestions on removing the old AS5 between the
CPU and the old Intel HSF? I've thought about using rubbing alcohol, but
I can't find it pure enough so as to not leave a film.

Then again, I will be using the same AS5 I used before, so perhaps it
matters less. I have a Teflon knife, which I could use to scrap off the
excess, maybe wipe it down with a lint free cloth, then reapply.

 

[NEM]

Hi Larry Gagnon,

I wouldn't get too anal about it. Studies show that expensive thermal
greases are rarely an improvement on regular inexpensive thermal heat
sink compound which you can get at any electronics store for next to
nothing. We have been thoroughly marketed to!

That might be true to some extent, but I already have the AS5 and
Zalmans' own (think Newmans' own salid dressing <G>) white thermal
compound. The time for marketing is past me, it's now just a choice
between the two that I have.

At best you might get a 2 deg C improvement. Big deal!

Cool (no pun intended), that will put me back in the 50 Centigrade

I would pre-spread the paste. Then you know for certain its thin layer
completely covers the die.

I'm concerned about bubbles, so I suppose putting it only on the CPU
would be best. Is that correct?

What about removal of the old AS5 from the CPU spreader?

 

[NEM]

Hi DaveW,

The Arctic Silver is far better than white thermal grease, and, yes,
spread it around.

But only apply it to the CPU, correct?

 

[Homer J. Simpson]

Why don't you just follow Arctic Silver's application and removal
instructions here:

http://www.arcticsilver.com/arctic_silver_instructions_big2.htm

 

[Ben Pope]

It's impossible to try all thermal pastes, However I do know the results of
a 12C difference on exactly the same Processor and heatsink between standard
retail silver paste and the Artic silver 5 makes a huge statement for how
superior it is. especially since the 55C reading was just before shutting
down the system, then puling the heatsink and processor, cleaning both with
rubbing alchohol, applying AS5 and reinstalling the same processor and
heatsink, then powering it back up and letting it idle for 1/2 hour for a
temperature reading.. with a 43C reading a 12C drop from only changing the
thermal paste.

So you changed:

- The installation method (assuming you hadn't just installed the
generic stuff).

- The amount of time that the paste had been on.

- The amount of time the machine had been run.

- The number of times you'd unsettled the paste by moving the machine.

- Probobaly some other things like fluff cleaning.

And you can decisively say that 12°C is the improvemnt in only the
thermal compound?

Ben

 

[Chuck]

Hi DaveW,



But only apply it to the CPU, correct?

--

No, to both cpu and the heatsink.

It's the AS5 on both mixing together that provides the themal link that
helps disipate the heat. Something should be clarified, thermal compound in
itself does nothing to lower CPU temps, thats the job of the HSF the thermal
compound simply makes it easier for the HSF to transfer heat from the
processor to the heatsink by filling in all the little grooves and
imperfections on the surfaces. I should add my 3200+ Barton is currently
running at 33C with a case temp of 32C with a room temp of 78F.

 

[Leythos]

And you can decisively say that 12°C is the improvemnt in only the
thermal compound?

I've been doing systems since before the 8086 was invented and using
heat-sinks for several decades. It's my experience that most of the HSP is
about the same across the board, that it's almost always a difference in
installation procedures that cause variances. Most people don't have a
clue as to how to apply HSP.

 

[Chuck]

I've been doing systems since before the 8086 was invented and using
heat-sinks for several decades. It's my experience that most of the HSP is
about the same across the board, that it's almost always a difference in
installation procedures that cause variances. Most people don't have a
clue as to how to apply HSP.

True, in fact thermal paste is definately a case where less is more, use as
little as possible and get it as thin as possible on both surfaces. too
thick and the thermal paste can act as an insulator, causing the heatsink
not to work as well as it should.

 

[signmeuptoo]

Can't be true, too many sites have tested products such as the Arctic
Silver and Shin-Etsu MicroSi products to be vastly superior when installed
in accordance with proper procedure as posted on the Arctic Silver website
and when no fingerprints or other contamination exists.

High performance thermal materials often improve over a curing time where
they form a type of structural and chemical bond between the CPU and HSF
surfaces.

One can argue all that they want that all thermal compounds are the same,
but too many sites (sorry, I have read them but I have not kept the links
to them, do your own research) have "proven" AS and S-E M to be superior,
and worth the tiny expenditure that they require.

You have to understand that products like Shin-Etsu and Arctic Silver are
designed by teams of scientific minds using chemical and other engineering
to accomplish the ideal particle composition, particle size, particle
variation, and so on. Too many Overclockers are proving to themselves that
better results in thermal conductivity can be had through AS or S-E for me
to believe that it doesn't make a difference.

ALSO, Cheap thermal compounds like you find at Radio Shack tend to dry up
and turn crumbly and cannot hold up to the temps that today's CPUs pass
out. So, IMHO, you cannot tell me that I shouldn't use my syringes of both
Shin-Etsu MicroSi and Arctic Silver Ceramique. They didn't cost that much
and they work fabulous in numerous ways.

 

[signmeuptoo]

Then, Larry, why have I seen so many sites revealing at least 12 degree
Centigrade improvements over cheap off the shelf thermal compounds? There
IS a big difference. The advanced thermal compounds are made by scientists
and engineers that know what they are doing, not by marketing experts.
They do cool things such as using several different particle sizes so that
there is a maximum surface area contact for best thermal transfer. They
also use much more chemically and mechanically advanced materials. The
goop that they use to hold the suspension is more advanced, the suspension
material is engineered through lots of research to hold up to high heats,
long use, etc. etc.

For the tiny increase in money for good, trustworthy thermal compound, it
is insane to not use it. Many years ago I made the mistake of using stuff
that I got from Radio Crap Shack. I lost a CPU because of it, it dried up
and crubled (the thermal compound), and I would have been better off if I
had not used ANY compound back then.

Larry, it is beyond me how someone who knows computers would advise against
such a cheap but wise investment in protecting HUNDREDS of dollars in
computer hardware.

OTOH, I, personally, would NOT use Arctic Silver 5. I don't recommend it,
because all my research indicates, based on user testimony in several
forums, that after 6 months or so, it cures so much so that the bond is
really strong, and removal of the heat sink from the CPU can be damaging if
a person is not really careful, smart, uses solvent (Acetone), etc. Arctic
Silver Ceramique is a safer bet. The only problem with Ceramique is that
it takes several days to "cure" and it improves in conductivity over that
period of time more so that the other thermal compounds. I don't know if
Shin-Etsu MicroSi cures like glue too, but I don't think so.

I, when I finally have the bucks and CAN build my system, will use my
Shin-Etsu MicroSi compound for the CPU and my AS Ceramique for the GPU,
RAM, Northbridge.

You can do what you want, but, at around 8 bucks, I don't think that it is
worth risking not using a quality thermal compound, besides, there is more
than one application when you purchase this stuff, so 8 bucks is like
nothing.

Some people used to say that high fidelity audio systems were just a
marketing ploy too, back in the 60's and 70's. They were wrong. So is not
using top notch thermal compound.

Sorry to argue I don't mean to be unpleasant, I just disagree strongly.

 

[signmeuptoo]

Homer, thanks for saying what I was about to. Follow the AS instructions,
TO THE LETTER!

Chuck, spread it, but don't apply it in any way that would let any of it
drip where electrical connections are because then it will cause a short.
Use a VERY thin application, and YES, transferring lots of heat is great,
but won't make much difference if a stock heat sink is used. It is using a
high performance heat sink-fan paired *with* AS or S-E that makes the
difference.

Here is part of a reply that I got from the people at Shin-Etsu MicroSi
about use of their compound:

"It is very difficult to say which product you have in hand.
As for cleaning, we can suggest Toluene, Acetone and IPA.
As for newer products, we have an MOQ of 50-75units.

As for how often you should change the material, I would say once you
notice a change in the performance of your computer, it might be one of
the areas to consider."

I had asked them which compound I got because it didn't have a label on it.
So they recommend Toluene, Acetone, and IPA (what that is I don't know) to
thouroughly clean the stuff.

Hope that I have helped.

 

[Chuck]

Homer, thanks for saying what I was about to. Follow the AS instructions,
TO THE LETTER!

Chuck, spread it, but don't apply it in any way that would let any of it
drip where electrical connections are because then it will cause a short.
Use a VERY thin application, and YES, transferring lots of heat is great,
but won't make much difference if a stock heat sink is used. It is using
a
high performance heat sink-fan paired *with* AS or S-E that makes the
difference.

Very true I have a Thermaltake Volcano 10+ Heatsink (all copper) with the
stock fan removed (was 70mm 10 mm maybe 20- 25 cfm ran at 3000 rpm) replaced
with a 70mm x 25 mm 7000 rpm 45 cfm Fan. And of course AS5, so it keeps
itself nice and cool. In fact right now it's at 29C, with MB temp of 28C and
a case temp of 75.5 F (24C). Not bad for a 3200+ Barton.

 

[Leythos]

Can't be true, too many sites have tested products such as the Arctic
Silver and Shin-Etsu MicroSi products to be vastly superior when installed
in accordance with proper procedure as posted on the Arctic Silver website
and when no fingerprints or other contamination exists.

High performance thermal materials often improve over a curing time where
they form a type of structural and chemical bond between the CPU and HSF
surfaces.

One can argue all that they want that all thermal compounds are the same,
but too many sites (sorry, I have read them but I have not kept the links
to them, do your own research) have "proven" AS and S-E M to be superior,
and worth the tiny expenditure that they require.

Actually, since most of them don't dry out in the first year, it's a close
call for most testing. Sure, they lose some of their gunk, but as thin as
the layer is (as it should be) it's not that big a difference. Don't
believe everything you read, do some testing yourself - I have.

You have to understand that products like Shin-Etsu and Arctic Silver are
designed by teams of scientific minds using chemical and other engineering
to accomplish the ideal particle composition, particle size, particle
variation, and so on. Too many Overclockers are proving to themselves that
better results in thermal conductivity can be had through AS or S-E for me
to believe that it doesn't make a difference.

I love AS, and I use it a lot, also use the stuff that Intel sends with
their CPU's in the syringe, it more about properly applying it than the
small differences in the composition at the two ends - meaning that cheap
is always low quality and expensive is always high quality, and that cheap
means less heat transfer and more expensive means better transfer. Now, if
you have some nice AS5 and apply it wrong it could yield worse results
than properly applied cheap HSP.

ALSO, Cheap thermal compounds like you find at Radio Shack tend to dry up
and turn crumbly and cannot hold up to the temps that today's CPUs pass
out. So, IMHO, you cannot tell me that I shouldn't use my syringes of both
Shin-Etsu MicroSi and Arctic Silver Ceramique. They didn't cost that much
and they work fabulous in numerous ways.

I never said you should or should not use anything, in fact, if you read
my post, you would see that I said that most of them are the same and that
the only problem is people not applying it properly.

Oh, they only work in one way - to create a better conduction of heat
between the CPU and the Heat Sink. Thinner is much better as it really
only needs to fill the micro-grain differences in the machined surfaces,
if the HS has a uneven surface you should refinish it or get a new one.

 

[Leythos]

why have I seen so many sites revealing at least 12 degree
Centigrade improvements over cheap off the shelf thermal compounds? There
IS a big difference.

Because there are two ends to the market - there is the cheap stuff and
the expensive stuff, cheaper means less heat xfer (in general) and
expensive means more heat xfer.

Improperly applied, the best HSP on the market is not as effective as the
cheapest HSP that's properly applied.