Slashdot Mirror
The Thermal Paste Revolution
arhines writes "ZZZ is running an article about an interesting new thermal paste which surpasses even solder in thermal conductance by 33 percent. If this paste makes it to the market sometime soon, we'll all surely be thinking about putting it in our boxes. In fact, if use of the paste becomes commonplace, it may even give the semiconductor industry a little speed boost."
Any of you who have ever even thought about overclocking are probably familiar with thermal paste and its function. Lots of OEM or low end cooling setups use either a thermal interface pad (TIM) or that white goop you get at radio shack. The fact is that neither of those does a great job of transferring heat from the processor to the heatsink. While they work ok, they don't exactly assist Moore's law in fulfilling itself by limiting clock speeds with heat. Yeah, CPUs are still getting faster, but one needs only to look at overclocking results with stock cooling versus those achieved with a good heatsink and good thermal paste to realize that better cooling = faster computers. Heck, if we'd stuck with the tiny old anodized fanless heatsinks on 486es, we might not be past 1GHz yet.
.1 micron diameter carbon filament was also tested. One would think that the carbon filament, being only .1 micron thickness would perform better - but this is not the case. The reason, Dr. Chung believes, is because of the porosity and compressibility of carbon black. Porosity plays a big role because wh
Current high end pastes range widely in composition, but in terms of performance they all fall within a very small degree range. Arctic Silver has been a longtime favorite among many, simply because they were the first company to release a competitive paste - one which was actually well suited to the task of transferring heat. Nanotherm is another big name these days, and lots of people are talking about PCM+, their upcoming metal-free product. But as I said, all of these products still get very similar results. A degree or two at most is all you can hope for in moving from one brand to another. So you might resign yourself to defeat saying, "Pastes have achieved perfection, so the bottleneck must now be the heatsink and the die of the CPU itself." And that would seem to me a very reasonable thing to say...the fact that several companies are putting lots of resources into the development of more efficient thermal transfer and seeing diminishing returns is fairly strong evidence to support such a statement. But according to materials engineer Dr. Junis Chung, current thermal pastes are rubbish, hooey, and applesauce, and he has something much better.
The paste he has created is based on dispersed carbon black (a particulate form of industrial carbon used to reinforce rubber) mixed in a soup of ethyl cellulose and polyethylene glycol. In tests comparing it to solder (a method of thermal transfer not typically used with electronic components because of the temperature required to bond it to both surfaces), the carbon paste surpassed the pure metal bond in thermal conductivity by 33%, and surpassed the taco-sntotting bond by 45% It was also superior to diamond and carbon nanotube based pastes currently undergoing development. Even if the carbon paste were to merely match the diamond and nanotube pastes, it would be a significant improvement because of the cost differences.
Why does it work so well? Spreadability. The problem with current commercial pastes is that they have focused too long on the thermal conductivity of the material, and not on the fundamental principle of a thermal paste, which is gap filling. Silicone based 'goop' from radio shack is fairly thermally conductive, but the size of the particles and the terrible spreadability cause it to be more of an insulator than a conductor. On the other hand, using something entirely liquid such as mineral oil doesn't cool well either because it isn't conductive enough, so the key is to find something with the right balance of conductivity and spreadability. The diamond paste contained particles 25 microns wide, about as small as they can be ground down to. This is what keeps the incredibly conductive material from winning - the diamonds actually end up pushing the two surfaces apart. To check for the importance of particle size in the opposite direction (smaller than the carbon black particles), a paste based on
according to the article you dont see the 33% improvement until you apply about twice the normal pressure. imagine the cracked cores
bite my glorious golden ass.
In fact, it was ( /.ed to hell now ) one of the more techy-nerdy-geeky sites that I've seen for quite awhile. I'd love to see more, but now I'll have to wait for 2 days.
I'm a subscriber - I pay for my right to bitch about /.
Have a look at the instructions for Arctic Silver 3 to see what kind of steps are needed.
Of course we should not forget that Using thermal grease on your Athlon will void your warranty.
Im dreaming ofa big bndwdth, That can resist the
The article on thermal paste is quite interesting, such advances are sure to aid in the cooling levels for the near future. However, I often wonder why there hasn't been much development in the way of devising viable (read: cheap) alternative cooling solutions (e.g water/fliud, air piping, effective passive cooling). It seems that the now archaic heatsink/fan just isn't cutting it anymore, at least down to a bearable level that is (the amount of noise my cpu fan creates is ridiculous).
Part of this is the chip maker's fault, for running the chip too hot/fast. Likewise, part of the fault rests on the case/fan manufacturers, as the cases become increasingly smaller, dealing with cooling becomes harder, as there is less space to work in.
It is getting to the point where I feel that my peace of mind with regards to noise is well worth the sacrifice of speed. After all, I don't need the full power of my cpu most of the time, just when compiling/rendering/encoding. The cpu just isn't the bottleneck anymore, and it's useless to continue in this speed race, not until the other system components catch up. Why doesn't the industry work together to create a better solution? It's high time I'm rid of the constant roar of these machines.
the net amount of heat generated by the processor will remain the same, so I'd suggest moving either the box or your legs to solve that one...
People replying to my sig annoy me. That's why I change it all the time.
* who runs within 2 degrees of max temp for their CPU? some crazy overclockers, but it's not exactly reliable practice, is it? if it was 10 degrees, maybe but it's not going to make that much difference
* stop knocking the thermal pads. retail CPUs use these because joe sixpack can't f*vck it up and claim on their warranty. if you don't like it, scrape it off and stick a blob of arctic silver or similar
* bear in mind AMDs warranty only applies if you use approved thermal solutions
This may help solve the problem that thermal compound applied badly is worse (in terms of temperature) than none at all.
In a thermal compound we are seeking somethng that:
(1) will conduct heat to the heatsink better than air
(2) will remain inert under extended high temperature exposure
(3) is non toxic (nice seeing as we have to deal with the stuff)
It is difficult for a material to conduct heat better than air if (large or many) air bubbles are present between the two surfaces, trapped by the compound itself.
So we all know how silicone performs, it meets 2 and 3 but there are some issues with 1, mainly because of the air bubble issue.
Carbon black, polyehtylene glycol and ethyl cellulose are both non-hazardous and ethyl cellulose is only mildly hazardous (Material Safety Data Sheets www.merck.co.th, criterion 3 met)
Particulate size is small (should lick the air bubble problem).
Spreadability should be a-ok (ethyl cellulose is a molding compound.
No polymerisation or other chemical reaction should occur (stable mixture, criterion 2 met).
Carbon is a brilliant conductor in this form ( criterion 1 met)
I think it'll work
Copper is a good heat conductor, and that's why it's used to spread the heat to a large surface area. The thermal paste has a different job: It only has to thermally connect the heatsink to the processor. It does not need to conduct heat over comparatively big distances. The kind of gap-filler between the heatsink and the processor is not as important as makers of the only substance which costs more per milligram than inkjet ink would like you to believe (at least as long as it's not air). The gap which the paste has to bridge simply isn't big enough for the paste to make a difference, compared to the thickness of both the silicon processor layers and the heatsink.
It actually will not help much. I've used both generic and that silver based Artic Gold (can't remember if that's the brand or not). I use MotherBoard Monitor to monitor temps on my XP/Game system which is equiped with a SLK-800 heatsink and 80mm fan.
There is a 0 degree difference between using the sivler stuff and the generic goo. I've also swapped from the goo to silver paste on my old dual 700 when replaceing a processor. No measurable difference in heat/performance.
Your best bet with those DSL routers:
Find a good 486 heatsink/fan combo, mix a drop of silver compound with a very small drop of epoxy, then mount the sucker on your DSL router's CPU. Use a bench clamp or book (or some combination) to secure the heatsink/fan overnight while the compound hardens. The next morning, your DSL router should run nice and cool. Keep in mind, if you use too much epoxy in your mixture, that heatsink will not be coming off there. A lighter mix will result in something you can knock off there with the handle of a screwdriver if you ever need to get it off.
I've found old 486 sinks and fans are very handy at cooling off just about anything they'll fit on except for peltier solutions.
If the cost of $5 is prohibitive, check your closet for old computers and find your free parts there.
Yep. If you peeled off that pad first, you would have probably gotten another 2-3 degrees. The pad uses a "wax-like" substance (i.e. a solid a room temperature, but liquid when lightly heated). The problem is with pads is that once you heat them after being installed, the wax-like substance just imbedded itself into all the microscopic cracks and holes on the top of both your processor and heatsink (doing its job). But now it is there, it doesn't just come off when you peel off that pad. It is there, and there pretty much for good, acting as a barrier between your heatsink and CPU and any other better thermal compound you use aftward.
:) ). This will remove the outer layer of the heatsink metal as well as the microscopic cracks and holes on it, which will include your heat-pad substance.
Now, I didn't say it was there perminent, but it is close to it. You can get it off the heatsink, as you simply need to heat up the heatsink (a very hot hair-dryer will do this). Once you heat it up, you can start wiping the heatsink down with a cloth. Or you can lap your heatsink (use several grades of sandpaper to get a polished, flat, smooth surface, usually starting with 100-300 grit paper and working your way up to 1000-3000 grit paper, depending on how "anal" you are
The CPU is almost impossible to fully remove the heat-pad substance. You don't want to lap a modern day CPU, as all you will do is "create" microscopic cracks and holes. Modern CPU's are laser cut and pretty much perfectly flat. There are "some" cracks, but they are much smaller/finer then almost any sand paper you will ever find. You also risk damaging the CPU as the manufacturers now have traces and transistors located micrometers from the top of the CPU surface. Heating the CPU can easily damage it if you are not careful about how hot you let it get. So it is usually very dangerous for you to try to remove the substance from the CPU if you have not already done it several times (or don't mind wasting whatever you spend on that CPU when you need to go out and buy a new one).
We were all warned a long time ago that MS products sucked, remember the Magic 8 Ball said, "Outlook not so good"
> AFTER peeling off the pad?! you know your suppose to peel that off BEFORE you install the cpu, right?
Ah, no. You're supposed to pull off the bit of tape that keeps it from sticking to other stuff and then squish it down on the CPU. Per the manufacturer's instructions, kind of thing.
Sheesh, evil *and* a jerk. -- Jade
Thermal pastes are thermally conductive, not electrically so. In fact, you want them to be a damned good electrical insulator. If you'd read the article (I did several days ago- [H]ardOCP had coverage of this one earlier this week...) you'd have found out that carbon black, the substance in question, outpaces pretty much everything else (including diamond and nanotube based compounds in development) because it fills the gaps between the heatsink and the chip's heat spreader, etc. with thermally conductive materials better than anything else.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
I solved that by picking up some Artic Alumina epoxy. That way the epoxy is built in. Worked great on my video card hsf.
*Note: No research was done, Artic Alumina was what was easily available. I'm sure your favorite brand/flavor has an epoxy version, too.
jred
I'm not a mechanic but I play one in my garage...
Actually pressure on the heatsink, helps in better
heat conduction by compressing the lattice structure of the molecules in the material, leading to improved thermal conduction. It also
helps squeeze out the tiny air bubbles, or increase their effective density, increasing thermal conduction again. There is nothing that
the pressure could get if the heatsink is made an
integral part of the chip enclosure surface(without any bonding material). But by
using a cold pressed annealed copper/silver you
could get slightly better thermal conduction(again
purity of the material could play a role in this).
So, ideally, you would like to have an extremely
thin,extremely conducting(not electrically) film
on top of the chip surface, that grows out as a
heat sink(which gain has to be liquid cooled to
be effective).
I strongly reccomend that anyone thinking of using this material request an MSDS first.
We were evaluating some material like this and it turned out to be composed of 30% Class 1 carcinogen. Would you store PCBs in your home?
Actually, the thermal paster does not help in
the heat recovery process(that a fan does).
Amount of heat removed is equal to the amount
of air moved out. You actually do not see the
localized heating of the cpu chip(because it is
way down in the layer of silicon), what comes
to the surface is more or less uniformly spreadout. The thermal paste does not do much
except to reduce the thermal resistance in the
heasink/chip cap interface. Still you need a
big fan, if the heat to be carried out is large
(as with current cpu's). A good analogy is to
think of a motor pumping water out from a sump
to a reservoir above through a pipe. The pipe
here represents the paste. What use is having
a thicker pipe, if the toplevel reservoir could
only be drained slowly(amount of air moved out
represents the heat that is drained, which is
proportional to the fan's air moving capacity).
Hope you understand there is NO OBVIOUS BENEFIT!.