Yes. Thermal paste does thin out when heated. This is actually good. It helps thin out the amount on the core. You only want the paste to fill in any imperfections on the surface on the heatsink and core. If you had a perfectly lapped core and heatsink and had good enough pressure and they mated 100% flat, then thermal compound would not be needed. The only problem, this is pretty much impossible to do, so thermal compound of some type is needed.
Now the thermal conductivity is not that big of a deal. They are rated in thermal conductivity per mm, but you will likely use only 1/10th that thickness. Several people say they are getting much better results with the Actic Silver likely have a very poor mating surface, requiring the better thermal conductivity, or they did not have the heatsink seated 100% flat. Even a thermal pad works good enough if the heatsink seated correctly.
The COMPUTER is your FRIEND!
Happiness is manditory.
AMD Phenom II x4 945 3Ghz | ASUS M4A77TD | 2X WD 1TB SATA 2 hard drive | 2x2GB Corsair XMS3 | nVidia GeForce 8800 GTS | ATI TV Wonder Theater Pro 550 | Antec P-160 case | Antec 650w Earth Watts | LG Blu-ray Super Drive | LG DVD RW | Windows 7 Pro
Actually, the measurement of thermal compound performance that applies to thickness is thermal resistance. The standard specification is for a layer 0.001 inch thick with a contact area of 1 square inch.
The cores of modern CPUs from both AMD and Intel are much smaller than 1 square inch so the effective thermal resistance of a thermal compound will increase as the contact area decreases.
The main reason you see discrepancies in the performance differentials between different thermal compounds is not so much due to application as it is to measurement technique. If you read through the Arctic Silver reviews at our site, you will see that testers using Intel CPUs with motherboards capable of measuring the actual internal CPU core temperatures and testers using well-designed synthetic tests report significant performance difference between compounds. Testers measuring the CPU temperature external to the core, as must be done with current AMD processors, report much less difference between thermal compounds. The science and math tell us however, that there should be a greater difference between any two compounds on an AMD processor as it dissipates more power than a comparable Intel processor and the basic formula for the temperature drop across a thermal interface is Thermal Resistance X Power Dissipated.
The reasons for this measurement discrepancy are explained in this post below that I originally posted to a forum a couple of months ago:
Unfortunately, many people do not realize that most thermal compound comparisons on computers do not reflect the actual performance differences between any two different thermal compounds. This is especially true on all AMD systems and any Intel systems that read the CPU temperature with an in-socket thermistor.
As our web site and others have pointed out for 6 months or more, temperature change compression is a natural phenomenon of secondary flow path when there is "both" heat flow and thermal resistance from chip to ceramic surface. This is the normal/natural case and cannot be avoided.
Not only are in-socket thermistors in a secondary heat path, they also only measure an average of the CPU rear temperature (minor contact with thermistor surface) and the temperature of the air in the socket (major contact with thermistor surface) which is itself an average of the CPU rear temperature and the motherboard temperature.
With both of these factors contributing to temperature change compression, the differences between different cooling solutions, be it heatsinks or thermal compounds is minimized. In most situations, this compression can be as severe as 4 to 1 so a 4 degree difference in core temperature is only measured as a 1 degree difference.
Also, isopropyl alcohol and acetone will not begin to remove all of the existing compound on a heatsink. The fibers on a cloth or paper towel are far too large to fit into the microscopic valleys that the thermal compound fills so once a compound is applied or a thermal pad is melted, some of that compound or pad will be on that heatsink forever. Any subsequent compounds will be --at least partially -- applied over the earlier thermal material.
So if you have had compound A on the heatsink and you switch to compound B, you are not really comparing A to B, you are comparing A to a mixture of A and B. If B is better, you may see some improvement, but probably not as much as if you had put B on in the first place.
Heatsink engineers understand this. Several manufacturers of high-end heatsinks apply Arctic Silver to the mating surface of all the heatsinks they send out for review then wipe it off. (This is known as 'tinting' the metal) They know that if the heatsink is tested later with a lesser compound, it will have a bit of an advantage because the AS will still be filling in most of the microscopic valleys.
In the final analysis, it all comes down to basic math:
Arctic Silver Compound: 0.001" layer 0.16 square inch contact area = 0.018 C/W thermal resistance
High-quality Non-Silicone Zinc Oxide Compound: 0.001" layer 0.16 square inch contact area = 0.125 C/W thermal resistance
Difference: 0.107C/W (And this is the best of the zinc oxide compounds and a layer only 1/1000th of an inch thick.)
CPU power dissipation X 0.107 = difference in CPU core temperature. So if your CPU is dissipating 50 watts, the CPU core temperature difference will be 50 X 0.107 = 5.35C.
The fact that in-socket thermistors will not measure this large a difference shows that the method used by most motherboards to measure CPU temperature while useful for getting a general idea of whether your processor is too hot, is worthless for making valid comparisons of thermal compounds.
The same measurement compression problems that apply to in-socket measurements also apply to measurements taken with a thermistor pressed against the side of the core. In fact, it is possible for core side measurements to actually reflect the relative performance of two compounds backwards. This occurs when the thermal compound squeezed out from the mating joint touched the thermistor alongside the core. The thermal compound with the higher thermal conductivity, while keeping the CPU core cooler, conducts more heat to the thermistor so that it reports a higher temperature.
I lap my HS with 400, then 600 grit. One dab of AS on the core, then spread it over the surface with a business card carefully. Install the HS and leave it with no probs at all. In the event I have to remove the HS, I have noticed that AS does partially liquify but does not cause any probs.
AS has caused instabilities for me if it accidently gets smeared on any contacts. Alcohol and a swab to clean, then start again.
My temps have never been better. Even wilder to note: I have some mods I do to standard CoolerMaster 6H11 HS's that have dropped 2 deg average using the stock fan which I prefer because noise has become a primary issue with me.
I encourage your research wholeheartedly and will stay tuned, but the fact that it is HYPER-CONDUCTIVE and you have not conducted tests on standard HS for comparison leaves it still in a pre-mature state.
I know that you are asking for help here in testing. We would need fresh, identical chips...say a couple DURON 800's with identical cases etc. I have all that...How much stock will I get ?
"I cant drive...55 !!!" Dr. Foxx Am.D.
I'd even put a dab of pure liquid silver on my core if it was going to drop the temp enough for me to squeeze some more juice I'd just be reeal careful about putting it on! I'd probably squeeze a bead of caulk around the edge of the ceramic too, to make sure
But for now, the tube of AS2 I ordered should suffice! I was worried about the reapplication thing, because I was going to use it on my dad's computer too, and I won't be around to replace it. But if it's not a problem, then no worries!
to clean just about anything electronic: fill up squirt bottles with acetone, isopropyl alchohol and deoniozed water. Rinse with acetone. Rinse with Iso. Rinse with a lot of DI water. Repeat. Repeat. Repeat. Repeate the last 4 steps (including this one) a bunch more times. Eventually, it will be perfectly clean Your proc's been through this probably hungreds of times, if not thousands... Unfortunately, I don't have a ready source of DI *sigh*. I do have this can of NAPA/CRC electronics cleaner tho- is it safe to clean heatsinks with? Ingrediants list: "Methanol 67-56-1, n-Hexane 110-54-3, isohexane 107-83-5, petroleum distallates 64742-48-9 and carbon dioxide". If the petroleum distallates included are volitile, I don't see how it would leave and oils behind... what do those #s mean?
If a person has seen a microscopic view of a polished piece of metal you would see that it is not smooth by no means. Although touching and perfectly polished(mirrored), under a electron microscope it is going to look like the Rocky Mountains. This is where thermal paste is needed as those two highly polished surfaces are only going to be touching at something like 10%(might be off) of the given surface area. As you can now see the thermal conductance of the joint is going to matter and it increases drastically as the thickness of the material for the joint decreases.
[quote]I think it just might sell ol'man. The conductivity is risky indeed, but heck, what do you think people were saying about water-cooled systems a few years back? [quote]
This is where I think people should protect their chips buy applying a very thin layer of rubber cement over their bridges on there cpu's. This rubber cement would then easily be removed by isopropyl or some other type of alcohol if needed.
As with AS, a little dab will do ya
I have had this applied to my HS for almost a month now and I haven't seen any change in temps except a slight possible decrease becasue of the compression factor. Temps are always right at room temp. Although some of that may be because of my setup for air flow also. It doesn't matter what the overclock is, at idle it runs the same as default voltage temps. Ironic as it may sound. I had this same setup and used the other thermal compounds and my temps were around 2 to 3 degrees higher than ambient. Higher for zinc. I never approached room temp. With this as I mentioned room temp is the standard
I do have this can of NAPA/CRC electronics cleaner tho- is it safe to clean heatsinks with? Ingrediants list: "Methanol 67-56-1, n-Hexane 110-54-3, isohexane 107-83-5, petroleum distallates 64742-48-9 and carbon dioxide". If the petroleum distallates included are volitile, I don't see how it would leave and oils behind... what do those #s mean?
Those numbers are CAS numbers. They are what all chemicals are identified by. wouldn't clean any heatsinks with it till I do a check on what the BP of them are. Chances are they are higher alkanes heptane, octane, nonane etc.... But if not and they are oils instead then I wouldn't use it.
I always clean my chips and stuff with toluene stinky stuff but it works great.
Here's a thought: Silver has simply amazing thermal conductivity, and there are a lot of hardcore OC nuts that willingly plunk down LOTS of dough for any sort of edge. See: Vapochill. So I wonder: why no company out there has put out a SILVER heatsink? If you think about it, silver costs about $5 an ounce on the market, probably a little pricier to get the physical product. But how much does a typical heatsink weigh? Around a pound, so the cost of the silver would be ~ $80-$100. Factor in the costs of actually getting the physical silver, building the HS and a decent cut for the manufacturer, the price goes up to $150, maybe $200. I'd bet there's alot of people that would be willing to pay that kind of cash for a SILVER heatsink. Heck it's silver! Hey and you can sell it later if you find something better. Or put it on a necklace and give it your girlfriend or something..."This is to cool you down baby, cuz you're so HOT" (I apologize I know that was TERRIBLE )
I have a feeling there's some painfully obvious flaw that I'm missing, but for now I think it's a great idea.
<A HREF="http://"www.jtbaker.com/msds/h2379.htm"" TARGET=_blank>normal hexane</A> (oo, life threataning and super-flammable! glad I don't spray this all over my hands while I'm holding something :eek
Isohexane (couldn't find a data sheet)
Hydrotreated heavy naptha
if that means anything to you on the can it says "Quick drying formula leaves no residue."- should I trust it? I use it all the time on electronics... doesn't remove dielectric grease or corrosion though but gets the grit away pretty good! At any rate, doesn't sound like there are any oils in here- just volatile organic chemicals that should evaporate.
Wow, I did not know that, have any links to the company that makes it, or to some reviews? BTW ol'man, I think you're goop would be a hit if it does work better than artic silver. You just need to get some serious testing done and find out exactly what its limitations are. If it works great long term, you got yourself a nice lil' product to sell
Nevermind, I found a review( http://www6.tomshardware.com/cpu/01q1/010306/index.html ), and I'M SOLD! It's not made completely out of silver, the contact plate is, but the real magic is: IT'S INAUDIBLE. Woot - finally a HSF for the quiet freaks! now I need to find where to get one.
Although I can see that it would work better if the whole base was made from Ag instead of just the part that comes out. I have already discussed this alot and it is good to know that there are other people that think that way. I would say similar to the copper hed hog but instead of copper it needs to be silver. Silver is the way to go IMO. Silver interconnects in CPUs and the like. Silver/gold alloy MOBOs Silver pins. Aluminum has the highest specific heat though so it would seem best to make the fins for any HS out of that. You don't see to many radiators with copper fins and I believe that is the reason behind it. I have the facility to make Ag Heatsinks if any one is intersted also. I have brought it up before but to no avail.
Hey cal, I guess I was too late but did you read my method of producing a real quiet computer and still having overclocks.
In the below picture my cpu is apparently running below room temp.
Usually this would run 2 degrees above room temp with the average compound but since the addition of my thermal compound(mix) I have gotten these temps. Many people dispute that temps can run below room temp without a pelt and or water cooling. That is not the point, the point is before the application, temps were higher, now they are lower at idle and even load by the same margin. Possibly better at load
Thanks for the info ol'man. I did read the thread on your cooling methods, very impressive. Although I still don't believe you can go below ambient temps with only aircooling (form a physics standpoint, it goes against the laws of thermodynamics) but you can clearly get very, very close Wow, if you can make Ag heatsinks, why not make a SUPERSILVER™ combo, and sell Ag heatsinks along with your special silver goop?
That Silverado is very nice, but it seems it's pain to get them in the U.S. right now. I might wait until I go to Europe this December and try to pick one up.