Best thermal paste that last 3+ years with minimal degrading?

I have a Y570 that overheats a lot on cpu, occasionally gpu, and I can't replace the paste on my own. A shop told me 50 dollars so I will buy the paste on amazon or newegg, but am unsure of which to buy. It is possible they won't seat the heatsink properly back onto it right? So I am leaning towards mx-4. It last 8 years and seems to do pretty good with bad mounts.

2011 Thermal Compound Roundup – Results Compilation | Skinnee Labs

 

IC Diamond 24 is the longest lasting paste on the market atm. It isn't quite the best but it can last for a very long time because its got a really high filler loading. It's practically a dough.

 

Don't know about you but I used the grey thermal paste and it has worked fine for me and I did it for my P-7811FX and haven't noticed any more heat then before that might change when I push it more but any thermal paste beyond the white paste is better and I don't recall Diamond being any better then the Grey thermal paste I used some have said some of those metallic paste might contribute to shorts rather then remove heat.....??

 

ICD is a great paste but if you just use stock components it will yield no discernible heat transfer advantage. Unless at stock you particular machine is heat monger to start with that is.

ICD will as stated though last longer than most standard TIMs. I have shown it can last well over a year under extremes with proper application and proper hardware alignment and pressure.

For standard installs your idea of MX-4 should be fine.

 

I had ICD7 on my G73JH and it ran strong for over 3.5 years... I can't recommend anything else...

 

Unless if the laptop's heatsink assembly is poor quality and applies pressure unevenly:

http://forum.notebookreview.com/har...iamond-24-giveaway-reliability-survey-55.html

Yeah, my laptop's heatsink has a 20% contact with the CPU die. IC Diamond failed after a few months of usage, and so did Arctic MX4. Read page 55 and 56, those are photos of my laptop.

 

Don't waste time with MX-4, my own experience has been that it flat out fails after 30 hours of gaming, with a 3C increase in load temps. And from there it'll just keep getting worse the more you use it.

Might want to have a look into ShinEtsu pastes, especialy X23-7921-5. My Clevo doesn't have very good contact pressure on my GPUs and this stuff is absolutely golden for low pressure/poor contact situations. ShinEtsu pastes supposedly hold up really well against "pumping out", so longevity shouldn't be an issue either.

 

The only paste I've had trouble with long-term is the white crap in the packets. It dries into a caked powder, becoming an insulator.

I've seen AS5 go for like 6 years. It gets thick but you don't overheat. I've also used Ceramique long term like that. Of course the stuff that's on the Intel and AMD heatsinks gets used forever and that might be Shin Etsu of some sort.

 

So which one will last a while with poor mounting and do well because I am not sure how well the computer shop will do with it and can't spend 50 dollars again and again. Also, what about Indigo Xtreme? And I can spend up to 20 for both cpu and gpu paste combined.

 

The problem is, when someone else does it, there are no guaranties. If they know the proper method of applying ICD then that would be the best bet. Again it is the best bet as there is no way to know what you will end up with.

I had silver paste on my Samsung for better that 1.5 years. This was done by the reseller. I just gave in and did Liquid Cool Laboratory Ultra but I had ICD on hand in case the contact patches looked questionable. While the Ultra is conductive if the shop is a good one and can use this then do it. It dries but to a metal contact that will last a long time. Your heat pipe contact has to be copper, aluminum will oxidize.

In all you are asking one of those questions that has been covered by thousands. Every TIM is a gamble, some are better than others and some last longer that others with every mixture of the two metrics combined. In the end 75% of the effectiveness, if not more, depends on your unknown and that is the heat pumping system along with contact patch limitations. Possibly a good indication of performance is that the Vendors selling customized systems offer ICD TIM as an option at additional cost but rarely any other TIMs.

 

I always used arctic silver, how come no one recommends them anymore? -_-

Never had a problem with paste breakdown, usualy something has broken on it within 2 years warrantying it's removal.

 

Arctic silver is great if you have sufficient and even contact pressure.

For laptops... Well my laptop had a 20% contact so any runny thermal paste will fail very quickly. Even the thick IC Diamond only lasted for 4-5 months before the temperature rose by nearly 20C.

 

^that.

On a desktop, or a device that cools down at the most once a day, a high visc.. viscous.... a thin paste will work perfectly fine for years if the components are evenly seated and no air comes in between the contact spots. And you would want to do that to get the best heat conductivity possible, to avoid uneven temperatures when the components suddenly heat up, etc.

The problem is that if you actually measure it the thicker pastes really don't have all that much lower heat-conductivity at reasonably high temperatures, as long as it's below the boiling point (i.e., they work well up to being turned into crisp, which happens at a bit lower temperature than for the hyper-expensive ground unicorn horn brands. And they're also less effective at low temps, but that doesn't matter).

So since laptops generally don't run at 105 degrees constantly, and since they're typically cooled down and warmed up all the time - a thicker paste is preferable because otherwise the paste is going to cake very quickly. That's why some manufacturers use thermal pads - they consistently last longer, even though they're bound to eventually fail within two years or so, than the most expensive paste you can get. And they'll also give you better and more consistent cooling, as well as that the cooling is going to gradually become worse (rather than fail instantly when the paste cakes and it becomes insulation - which happens all at once for the thin pastes...).

In other words - whether you choose a thin or a thick paste depends on how responsive you want the heat conductivity to be, against how long you want it to last.

And you could probably make the point as well that with most laptop cooling, the problem with getting the heat out isn't actually the conductivity between the chips and the copper in the first place. It's the heatsinks and the fans instead.

I mean, you've seen the article someone made a while back when they compared arctic silver and a bunch of other brands, against peanut butter and thick chocolate paste? I did something similar with with this fat grease my grandfather used to put in the inlet valve in the two-stroke engine in his fishing boat. Did it just for the hell of it with one of the pentiums - the ones that actually got really, really hot.. Anyway - only difference you get between the types are different boiling points, which is important of course (it should be slightly higher than water..). And that the different kinds retain the initial viscosity for longer or shorter..

 

I'm just going to copypasta this from my other thread...

If you guys are in the mood for a bit of reading, check out this PDF from Dow that explains how different factors such as thermal conductivity, thermal resistance, and bond line thickness are all related and affect each other.

If you want a more in-depth reading, then have a go at this paper that really digs deep into the interplay of temperature vs thickness, thermal conductivity, and thermal resistance for solders TIMs The most surprising part to me was they showed that below a solder TIM thickness of 0.1mm, increasing thermal conductivity or decreasing thermal resistance yielded no net benefit on temperature. Obviously this study was done on solder based TIMs so I'm not sure if the results could be directly extrapolated to TIMs made with other base materials.

Then there are additional factors like spreadability and wetability to consider, especially in a low mount pressre, poor die contact environment. It gets complicated very very quickly.