The Official Thermal Interface Material thread

[R-T-B]

I think they are just trying to say that their thermal paste has souped-up radiative heat transfer compared to other pastes
But before i heard of them i never once thought about how radiative my pastes were lol
So definitely sceptical but also intrigued
Needless to say i bought it and will compare it to my current SYY-157 paste soon
edit: going from a previous reddit answer sounds like the paste is tuned for use with a copper cold plate/heatsink, i wonder if that means that all that radiative magic won't work with nickel plated ones?
Well got an answer back
View attachment 317018

At least they are responsive to good questions. Keep us in the loop re performance. I'd love to be proven wrong, honestly.

 

[Shrek]

My guess is that radiative transfer is negligable compared to conductive transfer (at the temperatures involved).

Take for example a house radiator, most of the transfer is by convection, not radiation.

 

[Phenos]

Just received my order of GD-2, it's nice sturdy paste.

 

[Bobaganoosh]

It sounds like a whole lot of marketing. Most things emit IR radiation when heat is transferred. Of course, if they got hot enough, they'd emit visible light too. If you think about a metal rod (like a screwdriver for example) and you pass a current through it, it will heat up. First there will be infrared light emitted, then eventually it will get so hot it visibly glows red, then orange, then white.

So while they may possibly be trying to say more...what they've said is that it heats up during operation and emits infrared energy, which I would bet could be claimed of literally any thermal paste...or toothpaste...or Vaseline...or well you get the idea.

 

[lexluthermiester]

Heat into light? Lolwut. My sketchometer just went off the charts.

Actually, that is both a real thing and VERY common. This is why infrared camera's work. Everything gives off heat as photons.

Ok so maybe its possible, but I am still skeptical its going to come from a previously unknown group that decided to make... thermal paste.

This is a known thing. Unknown group or not, the principle is sound.

That's energy into light, not heat necessarily, correct?

Also a known thing. Incandescent lights work by giving off a tone ton of photon emissions, most of it infrared/heat, but enough of it in the visible band that it has always been useful.

 

[R-T-B]

Actually, that is both a real thing and VERY common. This is why infrared camera's work. Everything gives off heat as photons.

Not so much as its not possible as I don't see how that'd be terribly useful in enclosed thermal paste, and as such it sounds like buzzword speak.

the principle is sound.

Yes. I'm skeptical they actually use it still, but as eager as any to be proven wrong with results.

 

[lexluthermiester]

Not so much as its not possible as I don't see how that'd be terribly useful in enclosed thermal paste, and as such it sounds like buzzword speak.


Yes. I'm skeptical they actually use it still, but as eager as any to be proven wrong with results.

How it functions in a thermal paste is very much open for debate, but the science behind the heat bleed photon effect is factual.

 

[R-T-B]

How it functions in a thermal paste is very much open for debate, but the science behind the heat bleed photon effect is factual.

Yeah, I remember the principal now from some of this cool aerospace ultrawhite paints. Pretty cool.

 

[Borc]

I also bought Wx-130 as well as AMeCh SGT-4. I will try it on my laptop hopefully next week.

 

[oiawshboiwsahnoidaws]

I received my WX-130 today and seems like a great thermal paste
Probably the thickest paste i had
Compared to SYY-157 paste that i applied 2 weeks ago temps were lower by ~3-4°C
Btw i have the newer (?) SYY157 wasn't viscious like i heard about it so not sure where that one even ranks
Also i have little to no experience comparing thermal pastes so take it with a grain of salt
CPU is 12700k with power limit of 150W, Cooler AK620 with single noctua A12x25 fan and i maxed out case and cpu fan for comparison
Program used to generate load was "y-cruncher.exe skip-warnings stress -D:950 -TL:900 SFT"
I had pc in idle for a short time before running load for 15mins
Ambient Temperature was the same according to my cheapo thermometer ~20°C

syy157 - Google Drive

docs.google.com

wx130 - Google Drive

docs.google.com

 

[Borc]

The newer SYY-157 is easier to spread, less viscous. It's still a very good paste however. Ambient temperature was similar in your test? I got my WX-130 today, testing follows on Saturday I hope.

 

[3DVCash]

Thermalright TFX is awesome .

I used it on my Omen 2019 (i7-9750h, rtx 2070 max-q), first time/try repasting an expensive laptop after years of having a low end pc. lowered my idle temps (speedshift 0, so all cores working at max) by 5°C and max temps by 10°-15°C. a little difficult to apply at first without heating it up, but after you are done, great core uniformity temps, and no pumpout effect

So glad seeing this get a shoutout on the first page! I'm in love with this stuff. Its performance is on par with Kryonaut in my experience but it doesn't dry out nearly as fast.

 

[oiawshboiwsahnoidaws]

The newer SYY-157 is easier to spread, less viscous. It's still a very good paste however. Ambient temperature was similar in your test? I got my WX-130 today, testing follows on Saturday I hope.

Nice im interested in your results and yep ambient temperature was same-ish with 0,2C difference between tests but i know the used thermometer isn't that accurate
From my quick testing its a great paste but now im really interested if there is any magic in it or just a good run of the mill paste lol
Guess igorslab could find that out...

 

[Borc]

Nice im interested in your results and yep ambient temperature was same-ish with 0,2C difference between tests but i know the used thermometer isn't that accurate
From my quick testing its a great paste but now im really interested if there is any magic in it or just a good run of the mill paste lol
Guess igorslab could find that out...

My paste is extremely dry, it's not really spreading. Is it the same for you? Not sure if it dried out or expected. I will heat it up and give it a try.

 

[oiawshboiwsahnoidaws]

My paste is extremely dry, it's not really spreading. Is it the same for you? Not sure if it dried out or expected. I will heat it up and give it a try.

I could get my paste out of the syringe with ease and had some slight stringiness to it looked okay to me
But im a pea method kind of guy didn't try spreading it but i can imagine being hard since its a thick paste
Btw did you get a spatula with the paste? i thought it would come with one but didn't get one... i want my free spatula!! lol
I was curious and looked for pastes with similiar (stated) density and viscosity closest match is kryonaut extreme
kryonaut: viscosity 130-180 Pas, density 3,76g/cm3
wx-130 : viscosity 175 Pas, density 3,8g/cm³

 

[freeagent]

SYY and pea method is probably not a good idea, unless you want to insulate your cpu. SYY is best when spread thin.

 

[oiawshboiwsahnoidaws]

Yea maybe not a good idea pea'ing a thick paste but my SYY version is quite runny for some reason
Anyone tried https://www.moddiy.com/products/Hon...er-Highly-Thermally-Conductive-PCM-Paste.html ?
Seems to be an optimized version of PTM7950... but damn kinda expensive
Im quity happy with wx-130... but i think i got hit by the thermal interface material testing addiction hahaha

 

[Borc]

I could get my paste out of the syringe with ease and had some slight stringiness to it looked okay to me
But im a pea method kind of guy didn't try spreading it but i can imagine being hard since its a thick paste
Btw did you get a spatula with the paste? i thought it would come with one but didn't get one... i want my free spatula!! lol
I was curious and looked for pastes with similiar (stated) density and viscosity closest match is kryonaut extreme
kryonaut: viscosity 130-180 Pas, density 3,76g/cm3
wx-130 : viscosity 175 Pas, density 3,8g/cm³

I heated it up in warm water and this worked spreading it nicely. On my second and third spreading it worked without heating it up. Maybe it's because the CPU was still warm from the previous benchmark run which helped or maybe only the initial paste was a little dried out. There was no spatula in the package, only the paste. I tried several times and figured out that spreading doesn't work well with this paste on my laptop. The 5 dot method gave me the best result.

I use Kooling Monster KOLD-01 for reference because this paste works excellent on my laptop, this is the benchmark.

OCTT 30 minutes, same ambient temp and same fan speed.

Kooling Monster KOLD-01= 83.75°C (82/86/84/83)
Maxwell Labs Wx-130= 86.5°C (85/88/87/86)

It is not as good in the initial temp test. Compared to many other pastes it is reasonably well though, there are many other well known pastes which are worse on this laptop. So there is no magic from this paste on my laptop. Maybe this paste works better with higher mounting pressure or maybe it has better longevity.

 

[lexluthermiester]

SYY and pea method is probably not a good idea, unless you want to insulate your cpu. SYY is best when spread thin.

You're the second person I've heard that from. It's why I really don't want to try it. How thin does it need to be?

 

[:D:D]

PK1 my favorite I guess.

Did mean to do some further testing with measuring case temp and temp above TIM but messed up soldering to 2 free Super IO pins. Using a larger thermistor rather than thermocouple.

My angle grinder job probably not up to Intel specs and should not be tried at home but seems to work.

Maybe I'll get around to sorting it out one of these days.

 

[freeagent]

You're the second person I've heard that from. It's why I really don't want to try it. How thin does it need to be?

That thin

Don’t be scared, the power of freeagent compels you

 

[Count von Schwalbe]

View attachment 317712View attachment 317713

That thin

Don’t be scared, the power of freeagent compels you

I can't read the CPU through it. You have failed.

 

[freeagent]

I can't read the CPU through it. You have failed.

I should have showed the after shot of me lifting the cooler, sometimes you can read the IHS imprint on the cooler too

I do have a shot of that in this thread I think..

 

[Belfaborac]

Zero experience with SYY, but that's significantly thicker than I've ever applied paste. I used to start with a layer like in your picture and then removed just about all of it with a rubber-gloved finger until there was only a thin film covering the IHS. After a test-fitting showed a good and even connection of course.

 

[lexluthermiester]

PK1 my favorite I guess.

Did mean to do some further testing with measuring case temp and temp above TIM but messed up soldering to 2 free Super IO pins. Using a larger thermistor rather than thermocouple.

My angle grinder job probably not up to Intel specs and should not be tried at home but seems to work.
View attachment 317704
Maybe I'll get around to sorting it out one of these days.

What is this? And why?

View attachment 317712View attachment 317713

That thin

Ok, fair enough. As you've likely seen from my adventures testing TIM's in the last few years, you know I also use the spread method and always as thin as possible. Thank You for showing an example!

Don’t be scared, the power of freeagent compels you