The Official Thermal Interface Material thread

[Noci]

Few oils oxidize at the temps computer CPUs them expose to.

Yup, temperature increasements in chemistry often work as a catalyst in chemical reactions (like oxydizing). I would not find it strange if that happens with certain components of a TIM also.
However, the applied TIM between a cold plate and IHS/Die hardly has any (direct and/or prolongued) exposure to Oxygen, only a little on the edges so I'm curious to see a substantiated explanation from a chemist regarding this.
For all I know it might be a way more complex mechanism at work here, perhaps the pump-out effect has impact, maybe even within the TIM due to different physics of the solid and liquid components of a TIM. Not even mentioning the variables of mounting pressure, layer thickness and so on.

For TIM's drying out in their storage device, I would say: product design error or too much compromise to the balance of components/packaging .

So last week I was pondering regarding the custom loop cleaning and upgrade of my cooling system with the new parts I already have laying around (the annual cleaning is waaay overdue ).
I seriously considering about purchasing an Alphcool Core 1 LT and combining it with with a ThermalGrizzly PhaseSheet PTM (as soon as the latter is available in the Aquatune webshop).

For my graphics card I'm still doubting, will I repaste with my trusted Arctic MX4 (which I have used to my satisfaction for way over a decade), or make the jump and go for a ThermalGrizzly Kryosheet to see if I can bring the hotspot temp down (Delta-T is now between 25 ~35 C under full load).

Note: My Ryzen 9 5900x gets hot enough but for my graphis card I think phase change material has no added value as the temps are mostly around 44 ~ 47 degrees Celsuis under full load, and the phase change starts at approximately 45 degrees, so it won't reach it's full potential. Thats why the choice for Kryosheet or to stay with MX4.

Above mentioned temps are with Arctic MX4 from a 20g syringe of over ten years old mind you , so I have doubts about some statements regarding aging of silicon based TIM's and are in my opinion overrated.
For example precipitation can (relatively) easily be solved by remixing. Oxidation however is more of an irreversable process, so in my book the jury is still out .

But in the end, that's just me trying the relativize, feel free to disagree .

 

[Nordic]

Helios pads performed worse than I expected. https://www.igorslab.de/en/5-phasen...vs-pcm5000-pcm8500-und-thermalright-heilos/4/

 

[FreedomEclipse]

Helios pads performed worse than I expected. https://www.igorslab.de/en/5-phasen...vs-pcm5000-pcm8500-und-thermalright-heilos/4/

That is disappointing. Its given me great results with my laptop. I was going to use it on my AM5 build and 6700XT.

 

[oiawshboiwsahnoidaws]

Helios pads performed worse than I expected. https://www.igorslab.de/en/5-phasen...vs-pcm5000-pcm8500-und-thermalright-heilos/4/

wow this is surprising, i would have thought heilos would be on par with ptm7950 and also odd that ptm7950sp paste is worse too

 

[Nordic]

That is disappointing. Its given me great results with my laptop. I was going to use it on my AM5 build and 6700XT.

wow this is surprising, i would have thought heilos would be on par with ptm7950 and also odd that ptm7950sp paste is worse too

Helios isn't bad. It should perform well. Not being as good as the best isn't a huge fault as long as it performs well.

 

[Borc]

It's not odd, it was expected. The paste is much worse and not that good, I tried it some time ago. Other people who tried told the same.

 

[oiawshboiwsahnoidaws]

I used the paste and found it to be on par with heilos but it seems i now need to get my hands on a genuine ptm7950 pad to compare ^^

 

[FreedomEclipse]

I mean if we go over to tomshardware's megatest and look at their results.






It seems to be that Heilos works great so long as you keep it to 250w or below. Anything above that and you'll basically cook it. Which means for 90-99% of users the Heilos will work absolutely fine.

Igorslab biggest gripe seems to be that Heilos is too thin. Maybe double stacking them will give better result?

 

[Nordic]

Igor will be testing thermal grizzly's phase change sheet next. I am eager to see those results.

 

[FreedomEclipse]

Speaking of double stacking. I have a fresh Heilos and part of one left over from doing my laptop which is enough to do my 7800X3D twice. I'll stack them and see what happens. Either it gets better than normal temps or gets squeezed out. If i do this it could mean i wont have any left for my 6700XT but we'll see what happens.

Will know more in a few days. I do have some fresh Noctua H2 coming. My syringe of it that I was going to use was from 2019 and it supposedly only has a shelf life of 3 years. Proper vintage that.

 

[freeagent]

I am using a Heilos sheet on my 4070Ti.. its pretty decent.

 

[KLiKzg]

I mean if we go over to tomshardware's megatest and look at their results.

View attachment 359723


View attachment 359722

It seems to be that Heilos works great so long as you keep it to 250w or below. Anything above that and you'll basically cook it. Which means for 90-99% of users the Heilos will work absolutely fine.

Igorslab biggest gripe seems to be that Heilos is too thin. Maybe double stacking them will give better result?

This can't be right, as I have used MX-4 before & now I use Kryonaut...difference under load (which is what I do with BOINC) is something like 5~8°C, depending of the outside room temp.!

 

[FreedomEclipse]

This can't be right, as I have used MX-4 before & now I use Kryonaut...difference under load (which is what I do with BOINC) is something like 5~8°C, depending of the outside room temp.!

MX-4 has gone through some changes in the last few years although they might not have been open and up front about it

 

[Vayra86]

I'm going down the PTM road as well... and found out there's a betterer more optimized 7950, but paste! Gonna try this instead of a pad. How did I arrive here... because a single tube will do multiple applications, and those pads ain't cheap, and I also was surprised that there is a more optimized version already. Probably less coarse.

PTM 7958. Will update with results on my 7900XT *in 3-4 weeks...

Honeywell PTM7958 SP Super Highly Thermally Conductive PCM Paste

Buy Honeywell PTM7958 SP Super Highly Thermally Conductive PCM Paste for $24.99 with Free Shipping Worldwide (In Stock)

www.moddiy.com

^ also a great source for this PTM stuff to avoid fakes. On Amazon I simply could not find anything Honeywell branded anymore

 

[Uns1gn3d_C0d3]

^ also a great source for this PTM stuff to avoid fakes. On Amazon I simply could not find anything Honeywell branded anymore

Ebay is no better. The only Honeywell branded pads ship from China. The other ones that show up in a search look like PTM7950, but come with accessories that would make one think they're buying a paste.

I dunno. I took a pass on those and just bought a couple of pads from Linus' website. Should be here today.

 

[FreedomEclipse]

I'm going down the PTM road as well... and found out there's a betterer more optimized 7950, but paste! Gonna try this instead of a pad. How did I arrive here... because a single tube will do multiple applications, and those pads ain't cheap, and I also was surprised that there is a more optimized version already. Probably less coarse.

PTM 7958. Will update with results on my 7900XT *in 3-4 weeks...

Honeywell PTM7958 SP Super Highly Thermally Conductive PCM Paste

Buy Honeywell PTM7958 SP Super Highly Thermally Conductive PCM Paste for $24.99 with Free Shipping Worldwide (In Stock)

www.moddiy.com

^ also a great source for this PTM stuff to avoid fakes. On Amazon I simply could not find anything Honeywell branded anymore

​

PTM7950SP was tested in Igorslab review posted by nordic. It didnt do very well.

 

[Vayra86]

​

PTM7950SP was tested in Igorslab review posted by nordic. It didnt do very well.

Yeah I saw that AFTER ordering today lol, so I'm curious what I'll see here... not seeing disaster reports on the interwebs about people applying them though. Good temps. All I'm looking for really is a small improvement to hotspot temps. I don't need much, 3-5 C less on that reading gives me the headroom to push this GPU harder even in summer ambient. Right now at full tilt I'm seeing 105C, but then I'm pushing 370W through it, which I normally never do. The delta is high then though, it can be up to 28C, so I reckon I'll get something out of this.

 

[80-watt Hamster]

Yeah I saw that AFTER ordering today lol, so I'm curious what I'll see here... not seeing disaster reports on the interwebs about people applying them though. Good temps. All I'm looking for really is a small improvement to hotspot temps. I don't need much, 3-5 C less on that reading gives me the headroom to push this GPU harder even in summer ambient. Right now at full tilt I'm seeing 105C, but then I'm pushing 370W through it, which I normally never do. The delta is high then though, it can be up to 28C, so I reckon I'll get something out of this.

Moddiy's description claims 7958 is an improved 7950, so I'm still interested to see how it works in your use case.

 

[Borc]

PTM7950SP is like 4°C worse than Dowsil TC-5550 on my small laptop CPU. It's not the worst paste, actually even this result is above average I would say but not close to a top paste. Basically what igorslab tested.

 

[DirtyDingusMcgee]

I've been using Prolimatech PK-3 for a good long while. 40 bucks for 30g tubes, less for smaller ones. I use a decent amount. Consistently good results across many different system configs, within a degree or two of basically any quality TIM. Lasts a good while. If nothing else, the marketing blurb they have on their website just kills me

"Like two hot lovers, a heatsink and the processor want to make full contact. But microscopic imperfections in their craftsmanship create small gap for air. These gaps act as thermal insulators, inhibiting the transfer of thermal energy from your components. PK-3 thermal compound is made up of specially designed nano particles, engineered to correct these imperfections. The nano particles work by filling in the small air gaps and allowing for peak transfer of heat away from your components."

 

[basco]

i use different paste from i think 20 years ago and don´t have much probs.
mx4 , evga frostbite (who remembers), scythe thermal elixer , zalman g2......

 

[Nordic]

I mean if we go over to tomshardware's megatest and look at their results.


It seems to be that Heilos works great so long as you keep it to 250w or below. Anything above that and you'll basically cook it. Which means for 90-99% of users the Heilos will work absolutely fine.

Igorslab biggest gripe seems to be that Heilos is too thin. Maybe double stacking them will give better result?

I wish Igor would put these all on the same graph rather than having separate graphs for pastes and phase change materials. PTM7950's results are so good it seems anomalous. Thermal conductivity is one thing while real world performance is different. 75% of pastes in Tom's Hardware's mega test perform well. Helios didn't perform as well as I expected in Igor's test but Igor is testing what's possible not real world performance.

As long as you aren't trying to cool your PC with Kroger Gun Protection toothpaste you are fine. As enthusiasts we want the best cooling possible. We spend extra money to get 1% better results. Meanwhile the difference between most pastes and phase change materials is within margin of error. In my own testing the Thermal Grizzly kryosheet gave my GPU 2.7% more performance than MX-4. Was that money I spent on the Kryosheet worthwhile for a mere 2.7%? Yes. I would do it again. It was fun for me to test.

 

[FreedomEclipse]

In any case. Its shaken me enough to get some NT-H2 just in case the Heilos doesnt work out .
I ordered some from amazon and went to pick it up today but it turns out I made a mistake and bought some H1 instead.

 

[JWNoctis]

Strange how no major reviewer has tested Thermalright TFX, while TF7 and TF8 seemed common. It appears to have mostly fallen off the radar, despite claimed superiority over the TF8.

I've been happy with it for repasting laptop and NUC and a desktop build, with no issues except for the horrifying thickness. Wonder what they put into it.

 

[FreedomEclipse]

Strange how no major reviewer has tested Thermalright TFX, while TF7 and TF8 seemed common. It appears to have mostly fallen off the radar, despite claimed superiority over the TF8.

I've been happy with it for repasting laptop and NUC and a desktop build, with no issues except for the horrifying thickness. Wonder what they put into it.

Cant find anything about TFX apart from multiple complaints about pump out. I think PTM pads are honestly the way forward.