A question about how to apply thermal paste.

Caxus

I've always heard and read that a paper thin layer is best but since the paste is really only there to fill microscopic pits and imperfections, wouldn't merely smearing a tiny bit on with a platic bag around your finger be even better? Paper thin almost seems too thick.

edit: Also, are antec silver and arctic silver basically the same or is arctic better? I had some of both from my last upgrade a few years ago but i was only able to find the antec.

m-oreilly

after trying the "popular" methods, the small amount in the center of the cpu
then attaching the hsf seems to be fine for paste type compounds like arctic
silver (which i would prefer over the antec). i am using this with great sucess:
http://www.frozencpu.com/thr-26.html
(i have been ordering it from germany, nice to finaly see it stateside)

GHoosdum

I wouldn't use the stuff you have laying around from before. The particles tend to seperate out over time, so you'd likely not have as high a quality of conduction as a new paste would provide. I'd spring for a new tube of AS5 if I were in your shoes.

The thickness of the paste depends upon the flatness of the heatsink and CPU core or heatspreader, thinner is OK if you've got lapped or very high quality surfaces, but the general rule of thumb of "paper thin" usually does the trick best overall.

muddocktor

m.oreilly wrote:

after trying the "popular" methods, the small amount in the center of the cpu
then attaching the hsf seems to be fine for paste type compounds like arctic
silver (which i would prefer over the antec). i am using this with great sucess:
http://www.frozencpu.com/thr-26.html
(i have been ordering it from germany, nice to finaly see it stateside)

Good luck with that crap, m.oreilly. I wouldn't touch that with a 10 foot pole. It is devestating to aluminum, even if some just accidently contacts aluminum. Plus it is toxic to boot and I've also read that is dries out over time, losing some of it's effectiveness.

m-oreilly

muddocktor wrote:

Good luck with that crap, m.oreilly. I wouldn't touch that with a 10 foot pole. It is devestating to aluminum, even if some just accidently contacts aluminum. Plus it is toxic to boot and I've also read that is dries out over time, losing some of it's effectiveness.

total FUD there, chicken little...yeah, don't be using aluminum "crap"...:p

Leonardo

Mudd, I've seen that German stuff advertised, but hadn't read anything about negative aspects of it. Got a link?

madmat

If the slightest bit contacts aluminum it breaks the oxide layer and exposes the raw aluminum to oxygen. Since raw aluminum oxidizes nearly instantly to form the oxide layer that protects it, anything that removes that layer causes the aluminum to just continuously oxidize until it disolves.

I wouldn't suggest using that stuff in your $300 Lian Li V2000 or any other case made from Al. Just one drop and you've got a new ever expanding vent hole

V-P

madmat wrote:

If the slightest bit contacts aluminum it breaks the oxide layer and exposes the raw aluminum to oxygen. Since raw aluminum oxidizes nearly instantly to form the oxide layer that protects it, anything that removes that layer causes the aluminum to just continuously oxidize until it disolves.

I wouldn't suggest using that stuff in your $300 Lian Li V2000 or any other case made from Al. Just one drop and you've got a new ever expanding vent hole

That's why they don't allow thermometers in airplanes. Nowadays, very few thermometers use mercury, so they're not dangerous, but the ones that do use mercury will dissolve the Alluminum alloy that composes most of the airplane. It takes VERY LITTLE time for it to dissolve. It takes less time for mercury to dissolve alluminum then for you to dissolve two tablespoons of sugar in water by stirring.

BTW, if you go to google videos and put in Thermal Pste, you'll find a few tutorial clips, not to mention a perfect guide on AS's website.

m-oreilly

now i get it...these are joke responses, right? i use a lian-li v2000...lol, holes in aeroplanes...you guys had me going for a second, lol.

V-P

m.oreilly wrote:

now i get it...these are joke responses, right? i use a lian-li v2000...lol, holes in aeroplanes...you guys had me going for a second, lol.

I was serious.................................................................................

m-oreilly

SCAR wrote:

I was serious.................................................................................

hahaha, you're good...:thumbup i'll have to borrow your response. do you mind?

V-P

m.oreilly wrote:

hahaha, you're good...:thumbup i'll have to borrow your response. do you mind?

Yes, I'd like royalties.;)

airbornflght

No, he is quite serious. Mercury has a quite quick and adverse reactions. It's an oxidation reaction. Mercury will turn aluminim into aluminum oxide very quickly. Which gives the appearance of the mercury eating through the aluminum.

I'm not totaly sure about how it exactly works, but last year in chem, we put a drop of mercury on some aluminum, and it oxidized it in about 1minute. The way I understand it, aluminum only oxidizes a very thin layer on top and thats it, but I suppose that the aluminum keeps eating through the first layer and letting it oxidize again and again.

m-oreilly

SCAR wrote:

Yes, I'd like royalties.;)

my people will call your people.

csimon

Personally I still spread a thin coat with those fake credit card edges ...the way Mortin showed us years ago.

V-P

m.oreilly wrote:

my people will call your people.

You'll be needing to scedule an appointment with my secretary. Thank You.

AuthorityAction

csimon wrote:

Personally I still spread a thin coat with those fake credit card edges ...the way Mortin showed us years ago.

I use a razor blade, but it's the same idea. It's always worked for me and seems to make the job a lot easier.

Caxus

I've always used a razor blade as well i was just wondering if even thinner than that would be even better.

tmh88

Most thermal paste manufacturers suggest to use a small pea sized amount, and then just smoothe it out with a razor blade like alot of the guys have been saying

Thrax

I used my library card.

Books? lawl.

m-oreilly

Thrax wrote:

I used my library card.

Books? lawl.

hehehe

muddocktor

OK Leo and m.oreilly, here's some links and pics to see what this stuff does to aluminum.

Here is a thread at the overclockers.com forums on this stuff and the guy that started the thread did an experiment on an old aluminum heatsink. He has links in that thread to the pics he took, but I'm going to see if I can link to them in this post.

Sorry for the large size of them, but I didn't want to shrink them so it is easier to see what happened. Also, here is another link on damage at frostytech on a Zalman heatsink.

If you want to mess with this stuff then have at it. Even if it works marginally better than AS5, it's not worth messing with to me. Anything that corrosive has got to be able to be absorbed through your skin too and I don't want to take the chance, personally.

m-oreilly

i don't use alumium heatsinks. the instructions for the compound are quite clear.
the person or persons using it on aluminum either have not read them, or have another agenda altogether. it is a superior product, with clear instructions and the appropriate warnings provided. you might also show how dangerous it is to suit up for some snorkeling, and jump out of an airplane (with or without the mercury damage ) for similar carnage...

m-oreilly

and yes, your pics are rediculously oversized (lol). who pays for the bandwith/storage here anyway?

muddocktor

Hey bud, those are hosted elsewhere, so WTF is it to you.

And do you have stock in that company or are you a payed shill for them?

If you bothered reading the linked thread, you would see that the guy did it as an experiment to see what happens when this stuff comes in contact with aluminum.

You have made your point (whatever it is?) that you like this crap. I am showing reasons why I don't think it's worth messing with.

BTW, do you have an MSDS sheet for this stuff by any chance? I just looked at what SVC has posted on their website and this stuff has Gallium, Indium, Rhodium and Stannous (Tin).

wikipedia wrote:

Gallium is a chemical element in the periodic table that has the symbol Ga and atomic number 31....Gallium also corrodes most other metals by diffusing into their metal lattice. It is very important to keep gallium away from any type of metal containers such as steel or aluminum.

While not considered toxic, the data about gallium is inconclusive. Some sources suggest that it may cause dermatitis from prolonged exposure; other tests have not caused a positive reaction. It will, however, stain your skin if you hold it in your bare hands.

wikipedia wrote:

Indium is a very soft, silvery-white true metal that has a bright lustre.....Both gallium and indium are able to wet glass.

Pure indium in metal form is considered non-toxic by most sources. This may not be the case with indium compounds: there is some unconfirmed evidence that suggests that indium has a low level of toxicity. However, in the welding and semiconductor industries, where indium exposure is relatively high, there have been no reports of any toxic side-effects. Other sources are more definite about indium compounds' toxicity - for example, the WebElements website states that "All indium compounds should be regarded as highly toxic. Indium compounds damage the heart, kidney, and liver, and may be teratogenic." For example, indium trichloride anhydrous (InCl3) is quite toxic, while indium phosphide (InP) is both toxic and a suspected carcinogen.

wikipedia wrote:

Rhodium is a chemical element in the periodic table that has the symbol Rh and atomic number 45. A rare silvery-white hard transition metal, rhodium is a member of the platinum group, is found in platinum ores and is used in alloys with platinum and as a catalyst. It is the most expensive precious metal.....

Compounds that contain rhodium are not often encountered by most people and should be considered to be highly toxic and carcinogenic. Rhodium compounds can stain human skin very strongly. This element plays no biological role in humans.

wikipedia wrote:

Tin is a chemical element in the periodic table that has the symbol Sn (L. Stannum) and atomic number 50. This silvery, malleable poor metal that is not easily oxidized in air and resists corrosion, is found in many alloys and is used to coat other metals to prevent corrosion. Tin is obtained chiefly from the mineral cassiterite, where it occurs as an oxide.

Health Affects of Tin

The small amount of tin that is found in canned foods is not harmful to humans, but tin is mainly applied in various organic substances. The organotin compounds are the most dangerous forms of tin for humans. Despite the dangers they are applied in a great number of industries, such as the paint industry and the plastic industry, and in agriculture through pesticides. The number of applications of organotin compounds is still increasing, despite the fact that we know the consequences of tin poisoning. The effects of organotin substances can vary. They depend upon the kind of substance that is present and the organism that is exposed to it. Triethyltin is the most dangerous organotin substance for humans. It has relatively short hydrogen bonds. When hydrogen bonds grow longer a tin substance will be less dangerous to human health. Humans can absorb organotin substances through food and breathing and through the skin. The uptake of organotin can cause acute effects as well as long-term effects.

Acute effects: Eye and skin irritations, Headaches, Stomachaches, Sickness and dizziness, Severe sweating, Breathlessness, Urination problems

Long-term effects: Depressions, Liver damage, Malfunctioning of immune systems, Chromosomal damage, Shortage of red blood cells, Brain damage (causing anger, sleeping disorders, forgetfulness and headaches)

Just looking at these elements that are used in this product makes me want to stay away from it, besides the fact that this stuff is corrosive as hell with certain metals. I work with hazardous substances at my job and I've made it my business to know what kind of product I'm handling by reading the MSDS sheets and look for substances that can cause harm.

If you want to use this stuff then go ahead, but I suggest that you at least wear rubber (preferably neoprene) gloves while handling it to minimise your chance of skin contact. As for me, I'll stick with AS5 for now as it's most dangerous substance is silver.

Here's a link to that liquid metal tim's MSDS sheet, for anyone who want to read it for themselves.

Thrax

I would not, in a million years, get that **** close to my computer.

Yes. I have read the directions.
Yes. I am aware of the product's requirements.
Yes. I know what to, and what not to put it on.

I would not get that **** close to my PC.

Leonardo

Mudd, thanks for the links and pictures. I've attached your pics as thumbnails. At a screen resolution of 1280X1024 the pictures as they were stretched the thread page out to the size of a footbal field. I've attached the fourth picture to this this.

No, I wouldn't use that stuff in my computer either. I am sure it could be applied safely, but I'm not interested. m.oreilly, don't take it personally that some of us aren't interested in your thermal paste of choice. If you like it and get good results with it, all the power to you!

m-oreilly

"If you bothered reading the linked thread..." i was refering to general usage in my previous reply. to each their own :thumbup

Caxus

Well i tested it last night. It seems that i was wrong. By smearing on the thinnest layer i could i actually raised my temps about 2-3 degrees C. When i put a more generous amount on the temps went back down.

Leonardo

Take a look at Arctic Silver's instruction page. The recommend different application methods depending on the size of the chip that needs a thermal interface.

Quick summary:

small chips - spread it out with razor blade or platstic bag over the fingers
large chips (a la CPU's with heatspreaders) - dot in the center
rough/uneven surfaces - require more paste
polished surfaces that mate completely flat - minimal paste

Concerning CPUs with heatspreaders: I've tried the 'dot' method as well very careful application using a razor blade. Both worked equally well. The reason Arctic Silver recommends using a small dot of paste, letting the heatsink distribute it, is to avoid micro-air pockets in the paste, which reduces thermal conductivity.

Caxus

Wow, that stuff about only putting a dot in the center and letting it spread on it's own is interesting. I've always been covering the whole surface. I think most people do. I'll have to try that tonight.