• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Theremal paste woes

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

DeeXGod

Registered
Joined
Jan 5, 2005
Location
FL
Ok I had some ol' White thermal paste that came with my xp-90 and my temps were around 35-36 idle and 42c max, but I have some arctic silver 5, so I decided to use it. Well my temps were higher 42-45 :/ so now going back to the Thermalright compound that came with my HS mow my temps are around 38-40idle. Is it the way I am putting it on??? Too much too little? can someone explain the proper way to put the compound on. Thanks :cool:
 
Actually there's detailed manual on arctic silver site. I'm only confused why did you get higher temps. I'd blame too much / too little or maybe bad cleaning.
I applied paste according to guide, got nice drop of around 13*C (but I have Barton)... For A64 and P4 instructions are similiar, but differ from Athlon XP
 
Did you remove the compound that came with the xp90? if so, with what?

To apply thermal compound:
squeeze out a small dab (the size of an uncooked rice grain) on the center of your heatspreader. Then install the heatsink. The pressure of the heatsink will cause the thermal compound to spread.
You do not need to cover the whole heatspreader.


WELCOME TO THE FORUM
 
there goes the saying " if it aint broke dont fix it" just let AS5 to settle first it takes a while. Like Vio 1 said put a small dab, install the HS slowlly and rotate The HS left to right just a bit with a bit of pressure but not too much and monitor your temp as the AS5 settles in. :cool:
 
Remember, with thermal paste: less is more. Just keep trying to apply the AS5 and you'll eventually get the hang of it.
 
Hmmm well actully over time it got higher I had the arctic silver on for a fe weeks maybe a month. My Major question is I notice alot of instructions are for olser AMD chips with the small core area. I have a AMD64 so the heatspreader is really big and a amount the size of a uncooked rice would not cover much space, so is that for this chip too? Thanks for the replys. :D
 
Yes, the half-grain of rice technique is applicable on A64's because the actual core that's under the heatspreader is much smaller than the area of the heatspreader itself. A half-grain of rice sized dallop of AS5 with more than cover the actual core of an A64.
 
After 3 years of wondering why I can't get AS3 to work as good as my favourite white goop, I've given up, I'm declaring it the winner and I'm just using good white goop from now on. It seems to me that if you've got a nicely lapped sink, that white goop can do a better job than any of the silver compounds, if you've got a rough sink, then the AS is better. So I say, if it works good for you, use it.

regards,

Road Warrior
 
IHS = 1 or 1 1/2 grains of rice worth

Bare Core: 1/2 grains of rice worth


it won't cover the entire IHS, it doesnt need to, it may, but it doesnt need to, it only needs to yeild thermal contact near the core, where most of the heat is transfered anyways
 
RoadWarrior said:
After 3 years of wondering why I can't get AS3 to work as good as my favourite white goop, I've given up, I'm declaring it the winner and I'm just using good white goop from now on. It seems to me that if you've got a nicely lapped sink, that white goop can do a better job than any of the silver compounds, if you've got a rough sink, then the AS is better. So I say, if it works good for you, use it.

regards,

Road Warrior

I had really nice lapped heatsinks and got good performance using Zalman paste (white goop), however the best results I had were with AS5 (layer that you cansee through). I haven't tried Ceramique so far but it seems to allow very thin layer. Pastes that were said to be good, but had pretty big "pieces" gave pretty bad results.
 
Last edited:
i dunno davveey, i used the white goop before and my load temp was like 55, now its 49 with AS5... are you sure you cleaned it off properly with rubbing alcohol, or goof off?
 
Yep. As you would imagine pastes containing larger particles are better
for rough surfaces. Where small particle pastes are best for smooth.
 
I am thinking then I may have been putting too much on :D I will again try when I get off my *** and do it.
 
Ven0m said:
I had really nice lapped heatsinks and got good performance using Zalman paste (white goop), however the best results I had were with AS5 (layer that you cansee through). I haven't tried Ceramique so far but it seems to allow very thin layer. Pastes that were said to be good, but had pretty big "pieces" gave pretty bad results.
Indeed, that's what I've never understood about the 'grain of rice' installation method.

Is that just for n00bs or something?

I mean, how could that possibly get a better cover/interface than using an x-acto knife to spread a hair-thin layer of AS5?

Or, DOES it? Have I been applying the goop to the IHS wrong?
 
XanderF said:
Indeed, that's what I've never understood about the 'grain of rice' installation method.

Is that just for n00bs or something?

I mean, how could that possibly get a better cover/interface than using an x-acto knife to spread a hair-thin layer of AS5?

Or, DOES it? Have I been applying the goop to the IHS wrong?
because the pressure of the heatsink spreads it, instead of the imperfect human hand.

I blame crappy temp. monitoring.
 
XanderF said:
Indeed, that's what I've never understood about the 'grain of rice' installation method.

Is that just for n00bs or something?

I mean, how could that possibly get a better cover/interface than using an x-acto knife to spread a hair-thin layer of AS5?

Or, DOES it? Have I been applying the goop to the IHS wrong?

Yeah, or at least not as effectively as you could. I've tried both method (except using a razor blade instead of x-acto knife) and the "rice" method worked better (actually I did an extremely thin line down my Barton die). I think it may be because tiny particles of foreign matter get onto the die from the knife/blade. Just my theory. That, and you usually end up putting a little too much on with that method and thicker in some places than others (this isn't visible to the naked eye). When you let the paste spread out on its own, it seems to fill the gaps more effectively than when you try to spread it out and then it spreads out further on its own. Just my 2/5 of a nickel.
 
thats what Nevin specifies in the directions

for a CPU with an IHS, you need 1 to 1 1/2 grains of rice of a BB-sized dab of it on the IHS

the science behind this is AS5's break-in period, part of that period is the spreading of the thermal compound to form a micro-thin layer of it across the IHS, and during this process, it will fill microscopic gaps in both the IHS and the HS base as it spreads, because of the way it is forced to spread out, its forced into these microscopic gaps, and that helps increase cooling capacity

plus, note that the core is not the size of the entire IHS, so you're just wasting compound if you spread it over the entire IHS

i used to do the razor trick like you, then i finally read the AS5 directions, and gave it a spin - my load temps dropped 8ºF from the razor spread technique after about 2 weeks worth of normal CPU use (plus the usual daily hour of P95 torture test - don't ask, its a habit :p)

load used to be 112ºF, now its 104ºF

the 104ºF is also with a GPU block added to my loop, along with DUAL P95 torture tests (the old 112ºF was done with a single instance of P95 TT)
 
Interesting. I may have to give that a try, then. Right now I'm using a Zalman CNPS7000-AlCu, but I'm *REALLY* not happy with the temps (41C idle, 54C load), so was looking at an XP-120. Might try the alternate mounting method with the new heatsink.
 
XanderF said:
Interesting. I may have to give that a try, then. Right now I'm using a Zalman CNPS7000-AlCu, but I'm *REALLY* not happy with the temps (41C idle, 54C load), so was looking at an XP-120. Might try the alternate mounting method with the new heatsink.

alternate??? :eh?: its the recommended method by the manufacturer! :drool:
 
Back