IBM Doubles CPU Cooling With Simple Change

Ars Technica is reporting that IBM has discovered a new cooling breakthrough that, unlike several other recent announcements, should be relatively easy and cost-effective to implement. "IBM's find addresses how thermal paste is typically spread between the face of a chip and the heat spreader that sits directly over the core. Overclockers already know how crucial it is to apply thermal paste the right way: too much, and it causes heat buildup. Too little, and it causes heat buildup. It has to be "just right," which is why IBM looked to find the best way to get the gooey stuff where it needs to be and in the right amount, and to make it significantly more efficient in the process."

Re:Excellent

[unborracho]

sure, it does. Less fans = less power consumption.

Re:And people thought they were cool polishing....

[planckscale]

I've read it's best not to polish but to use a very fine grain sandpaper to rough the surface up just right. And don't tell me I don't know how much paste to apply. I'm a proper paste amount applier thank you very much.

Re:So... what did they do?

[malfunct]

They etched a series of microgrooves on the surface of the headsink to act as a channel for excess thermal paste. This is supposed to make much better contact than a smooth surface.

Artic Silver provides great instructions...

[madhatter256]

When i ordered my Artic Silver compound, the website had some instructions on how to apply the paste depending on what type of CPU you own. These instructions can be applied to any kind of thermal paste.

here's a link.

http://www.arcticsilver.com/instructions.htm [arcticsilver.com]

Re:And people thought they were cool polishing....

[feceus]

If you'd ever taken the time to actually try lapping the heatsink and heat spreaders rather than making fun, you would notice a significant drop in temperatures.

Even today with the new Core 2 Duo CPUs, the IHS have been found to be concave. Personally having lapped my CPU, the load temperatures dropped 10 C - nothing to sneeze at.

This article is more about the refinement of a technique. Notice how the article states "micrometer-length trenches", and not surfaces filled with ridges you can feel by running along it with your finger nail.

Most overclockers know that you get diminishing returns the further you polish the surfaces anyway.

Re:Excellent

[Chacham]

It doesn't help power consumption, but better cooling = less fans = less noise.

Actually, it helps *very* much with power consumption. Usually, resistance goes up as the tempeature does. For example, this is what an incandescent bulb relies on. What this means, is that as the chip gets hotter, it will resist more, causing a need for higher output to get the same usuable energy. By cooling the chip, its resistance stays low, allowing a higher efficiency in power usuage. IOW, less heat, less energy required.

Secondly, as another commentor pointed out, there's the fans that are use to cool it down, which indirectly allows for a lower power-consumption.

Previously announced in October

[writertype]

Extreme Tech had this last year. With even more pictures! :)

Story is here [extremetech.com].

Re:Excellent

[Anonymous Coward]

Silicon transistors go the other way -- resistance decreases as temperature goes up.

Re:Excellent

[Schraegstrichpunkt]

Actually, it helps *very* much with power consumption. Usually, resistance goes up as the tempeature does. For example, this is what an incandescent bulb relies on.

IIRC, semiconductors don't work that way; Their resistance tends to decrease with increasing temperature.

Re:the last time i did it

[Anonymous Coward]

First, you should have done some research. I don't care how much was in that syringe, it was probably too much -- you only need a small amount of thermal paste, just enough to fill the tiny gaps between a CPU and the heat sink.

Second, removing the case's cover will completely disrupt the air flow inside. If that actually makes your CPU cooler, you have some serious problems with the way your fans are set up. If they're set up so that they're constantly pushing cold air over the CPU and hot air out of the case, it should, in fact, be cooler with the cover on.

Re:And people thought they were cool polishing....

[thrawn_aj]

along comes IBM and makes them rough and it cools better :)

Hmm, I am not a modder, but I am a lab rat and roughening is a common technique used to increase the effective available surface area that is in contact with the heat-sink compound. This is not limited to CPU cooling and it's a little strange that it's taken so long to implement. Chemists play the same trick when they want to increase the rate of a reaction, powder up your reagents, or your catalysts. Of course, this will work only up to the point where the heat-sink particles (micron sized here I'd guess) can SEE the extra surface area. Hence, there exists a limit to how rough or how fine you want the surface beyond which range the cooling gets less efficient. A fine grit sandpaper (as a responder suggests in this thread) should be the way to go.

Re:Did you read the article?

[maxume]

An easy way to think about it is that the paste is better than an air gap, but worse than contact.

Re:And people thought they were cool polishing....

[sdack]

It is less funny when you realize that the roughness stands in a direct relationship to the size of the metal bits in the paste. If all you can get between the valleys of a roughened copper heat sink is the binding mass instead of the silver particles because their too large then you will have a rather bad heat conduction. If you however get the surface rough and the silver particles are as small as nano particles, then you might get what IBM has achieved: much more surface and lots of contact.

Re:Good, it was the worse part!

[SatanicPuppy]

I did that once, and had problems, and then opened the machine up to find that they'd just squeezed a dime-sized glob of paste on the cpu, and then mashed the heat sink down on top of it...There was a visible gap between the CPU and the heatsink.

I "noticed" it almost immediately because of the massive increase in fan noise...the fan was supposed to be replaced with an identical fan, so I thought they'd just screwed me, but the fan was correct, so I checked the cpu, and voila, craptacularity.

The easiest way to apply thermal paste is to seat the processor, and then dump a glob of paste on it, and spread it as thin as you possibly can with a credit card or some other plastic scraping tool...Don't be afraid to scrape off the excess! That's what you're supposed to be doing! Then mash the heat sink down on it, and see if any squeezes out the sides...If it does, you've got too much. Scrape some more off. You should definitely be able to see the top of the CPU through the paste.

Easy as pie. The only time I ever saw anyone have problems was with one of those old Socket A AMD processors, where you had to have a fricking screwdriver to force the metal clamps on the heatsink into place...Lot of people put holes in their motherboards while installing one of those chips.

Re:the last time i did it

[drinkypoo]

Second, removing the case's cover will completely disrupt the air flow inside. If that actually makes your CPU cooler, you have some serious problems with the way your fans are set up. If they're set up so that they're constantly pushing cold air over the CPU and hot air out of the case, it should, in fact, be cooler with the cover on.

Most cheap PC cases are designed utterly without thought to proper airflow.

Also most times fans blow in from the front, across the drives, where the air is preheated.

Most cheap PC cases will cool better when open, sad but true.

Re:Did you read the article?

[Chirs]

High performance thermal pastes (Arctic Silver for instance) are too thick for that technique to work. You need to smear it evenly across the cpu before putting the heatsink on.

Re:Did you read the article?

[theantipop]

Not true. Artic Silver changed their recommended instructions [articsilver.com] a couple years ago to the BB-sized dot in the middle of the core technique. This reduces the chance of air bubbles that can occur when you try to level the compound manually because the pressure of applying the heatsink will do a much better job.

Re:Did you read the article?

[Anonymous Coward]

One thing the DIY Audio crowd has tried in connection with the reappearance of power JFETs (albeit in tiny packages with very little surface area to dissipate heat from, hence the relevance) is to use an indium based solder. These turn fluid at temperatures that the core should be able to withstand for a short amount of times. Hence, you put the heatsink on a stove, put a tiny bit of indium solder on it, and carefully heat it to the exactly right temperature and then apply the CPU. A solder interface is still worse than a single piece, but will be pretty nice, and indium actually wets the metal. Far better than a similar amount of grease.

Re:Did you read the article?

[Anonymous Coward]

According to Wikipedia, thermal paste has a thermal conductivity more than 100 times worse than copper.

http://en.wikipedia.org/wiki/List_of_thermal_condu ctivities [wikipedia.org]

Re:Sadly

[stmfreak]

This isn't taught anywhere.

At the systems company I worked for, we were told to use the entire tube as well. Granted, it wasn't 4ccs, but it was still too much. Our CPUs would typically have 1mm of paste between them and the heat spreader--easily seen when you took them apart later.

Back when I was overclocking my white-box PCs, I read that paste is only supposed to fill the grooves between heat sink and chip die. Ideally, you want metal to die contact, but since these surfaces are typically non-uniform, direct contact is very inefficient with only a few high-points touching. Therefore, thermal paste was developed that allowed the gaps to conduct a bit better than air. However, apply too much and you eliminate the possible metal to die contact and thus cause over heating.

I also read about "lapping" and actually did an experiment with one of my dual proc PCs. I sanded my Celeron die and heat sink flat with 400 grit sandpaper and then put them together with the smallest dab of paste. Separating the two revealed that nearly all of that paste was pushed out of the contact region leaving a very thin, nearly non-existent film of paste. Back in my system, this CPU was immediately 5C cooler than its twin, under any load.

Metal to die beats paste any day. These companies could do a lot better with focus on making flat surfaces, less viscous paste and a little education.

Re:And people thought they were cool polishing....

[Glonoinha]

Very true, but the issue I was getting at was that the thin layer of gold would be instead of the thermal grease, and would serve to create the 'gasket' between the CPU and heatsink, increasing (significantly, if my theory is right) the thermal transfer as the gold would have a much higher coefficient of thermal conductivity than even the best paste. The reason I suggested gold is that gold can easily be purchased in small quantities of gold foil, the gold foil is ~very~ flexible / malleable and would serve to fill in all the microscopic gaps between the CPU and heatsink creating a thermal bridge in the process (which is the purpose of the paste), thus making the cooling solution quite a bit more effective.

In theory.

Anybody want to try it, perhaps on some older hardware?

Re:And people thought they were cool polishing....

[binarysins]

Sounds like it would not work well:

"To be effective a TIM combines properties to minimize the total interface resistance. High conductivity (200-420 W/mC) materials, like copper, silver, aluminum and gold, maximize thermal conductivity, but do not flow into intimate contact because of the relative lack of compliance so the interface resistances are very high and the overall performance is poor."

http://www.indium.com/_dynamo/download.php?docid=3 10 [indium.com]

Re:Did you read the article?

[maxume]

If you google for "0.0045C-in2/Watt." it becomes clear that the number they are giving is for a layer 0.001 inches thick. That's barely there.

It also mentions that the stuff is 99.9% silver, which is dandy, but the difference between silver and copper is at the 'you aren't gonna notice' level:

http://en.wikipedia.org/wiki/List_of_thermal_condu ctivities [wikipedia.org]

If your application is such that money doesn't matter, silver is the obvious choice, but that's about it. A solid lump of something will generally have a better conductance than a paste, since it is 'stuck together' more, equal depths of copper metal and paste on an ice cube would demonstrate this.

A good application of a decent paste will outperform a bad application of a good paste, it just doesn't matter a huge amount.