[H] Thermal Compound TIM Round-up Candidates

Since Cheese has already been done we need another oddball this round! Maybe some elmers glue......
 
Excited to see this round up. I'm still alternating between AS5 and the Shin Etsu for the last 5+ years.
 
Great idea doing another TIM comparison! There are only a handful of viable ones done infrequently, to the point when the ubiquitous "what TIM should I use" questions arise, people often end up pointing to a comparison done several years back. Good idea avoiding "non-traditional, highly-fuckupable" products. Those TIMs are in a league of their own both with difficulty/complexity of install, electrical conductivity, and performance.

Seems like you have a good mix of products there. You've got AS5 and ShinEtsu, the best of the elder generation. A good selection of ArcticCooling , Tuniq , and Phobya solid performers. IC Diamond both "7" and "24" models, which have a long history too. Gelid GC Extreme and Prolimatech PK1 and PK3 are noted to be among the best of recent years, with ThermalGrizzly Kryonaut being the new champion last I checked; I'm curious to see if it defends its title and by how much. I'm curious about the Coollaboratory Liquid Copper, given that their previous product was a liquid metal in the "conductive and highly fuckupable" category, but it seems this one is a more traditional TIM. Looks like a good group!
 
Xaeos said:
Great idea doing another TIM comparison! There are only a handful of viable ones done infrequently, to the point when the ubiquitous "what TIM should I use" questions arise, people often end up pointing to a comparison done several years back. Good idea avoiding "non-traditional, highly-fuckupable" products. Those TIMs are in a league of their own both with difficulty/complexity of install, electrical conductivity, and performance.

Seems like you have a good mix of products there. You've got AS5 and ShinEtsu, the best of the elder generation. A good selection of ArcticCooling , Tuniq , and Phobya solid performers. IC Diamond both "7" and "24" models, which have a long history too. Gelid GC Extreme and Prolimatech PK1 and PK3 are noted to be among the best of recent years, with ThermalGrizzly Kryonaut being the new champion last I checked; I'm curious to see if it defends its title and by how much. I'm curious about the Coollaboratory Liquid Copper, given that their previous product was a liquid metal in the "conductive and highly fuckupable" category, but it seems this one is a more traditional TIM. Looks like a good group!
Click to expand...

IC Diamond "7" and "24" are the same exact thing. The only difference is the quantities. Instead of measuring them in volume, they are measured in weight.
 
Phanteks TIM has always served me well, and it's cheap. Maybe snag some of that. Also, chocolate.
 
Every experiment should have a control so how about including the plain old white silicone-based thermal grease that was used extensively before TIMs became more sophisticated? This would show how much TIMs have improved.

(Hopefully a lot!).
 
For an unusual TIM, has anyone ever tried Vaseline (aka petroleum jelly) ?
 
RedWagnum said:
So 7 karat and 24 karat? Thasa whole lotta diamond!! :LOL: :ROFLMAO: :woot: :D
Click to expand...

Heh.. there is more in it than just diamond dust.

And yeah, it is the weight.

7 carat = 1.4g (They should really have named it 7.5 since it is 1.5g, but that doesn't sound as fancy as 7)

24 carat = 4.8g
 
I'm also really interested in what would happen if you just take the IHS off, remove most of the factory glue, spread the TIM back out a little carefully and put it back on. We've all been speculating for a long time now that just reducing the gap/increasing the pressure on the die would improve temps a lot. Let's see that test done FIRST as a baseline before doing any of the other compounds.

I've been through a bunch of TIM over the years. For customer work I STILL use AS5. It has properties that are just plain superior. It's thick, but it's just the right thick. You can still apply it easily without crazy tools or heating it first. And it stays good delivering good temps for the life of an average system. A lot of the exotic stuff gets... funky. Once you get used to working with it, the application is actually preferable to the more liquid pastes. It goes right on and smooths right out and doesn't move. The only trick is that it really doesn't like oils on the surface of the chip or it gaps when spreading. But really, who doesn't spend 5 seconds cleaning with alcohol or other surface cleaner first?

That all said though, if I was doing a one-time delid job I'd just say screw it and put the metal TIM on. I keep one tube of it around for that kind of thing.
 
I always love the "Wow I used this paste and my temps dropped by a massive 15 degrees!"

Nah, your heatsink wasn't properly attached before!
 
Last edited:
Just in time. I need to order some and I'm on the fence between nth1 and gc extreme. Or maybe this will change my mind altogether!
 
Simliarly, I ran out of MX-3 a few weeks ago.
This will be helpful.
 
Although I'm sure the goal is to review some of the higher end product available on the market today.
I would be very interested in seeing how some more standard fare like mg chemical 860 or even some of their "premium" products like 8616 or 8617 would compare.
I personaly go through 1.5-2 tubes of mg 860 when de-lidding re-balling and re-lidding PS3 Cell and RSX Processors.

As far as oddball suggestions for shiggles (like you did with cheese) I have always wondered how some gold leaf would stack up against some of these premium pastes 5 sheets can be had for just $10 and It would be an eye opener.
 
Thermal Grizzly Kryonaut is in my top 3. Needs a little prep time in the hot water tank before application tho - it's thick.

As suggested previously, would like to see Thermal Grizzly Conductonaut compared too.

Instead of cheese, what about some thin Fujipoly Sarcon XR-m slices?
 
Since we're already going to be swapping TIMs, should we throw in some TIMs cleaners as well to see if they do what they claim any better than alcohol?

The two I constantly hear about are Arctic Silver's Arcticlean (http://a.co/0zypPzS) and Akasa AK-TC (http://amzn.eu/h6Uhlok but tot sold in US. Found it on ebay a few times.)
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
Might be worth pointing out how hard one is to clean and replace vs another. Or if they come off the same.
 
Qrash said:
Every experiment should have a control so how about including the plain old white silicone-based thermal grease that was used extensively before TIMs became more sophisticated? This would show how much TIMs have improved.
Click to expand...
Plain old white TIM?
I know Kyle loves [ H ]ardware, but not enough that I think he'll ...

 
I use 99% isopropyl alcohol for cleaner. Not covering other products.
 
Qrash said:
Every experiment should have a control so how about including the plain old white silicone-based thermal grease that was used extensively before TIMs became more sophisticated? This would show how much TIMs have improved.

(Hopefully a lot!).
Click to expand...

how about measuring without TIM as control point ?
(also, as persuasion: it can be found everywhere, it's not expensive, shipping is free and quick, and probably very easy to clean)
 
Qrash said:
Every experiment should have a control so how about including the plain old white silicone-based thermal grease that was used extensively before TIMs became more sophisticated? This would show how much TIMs have improved.

(Hopefully a lot!).
Click to expand...
Yes, will be doing this as we always have.

bl4C3y3 said:
how about measuring without TIM as control point ?
(also, as persuasion: it can be found everywhere, it's not expensive, shipping is free and quick, and probably very easy to clean)
Click to expand...
Yes, intend on doing this as well. Not sure if it will make it through a load test or not.
 
You should try out the graphene paste. I've got a tube from these guys (it was a promo so I'm not sure if they still sell it in "consumer" sizes) and it's worked well for me. I haven't had enough time to test all the variables, but I did see a massive drop when I switched from my standard as5. I'd love to know if that was just improper mounting the first time, or if this stuff has merit.

https://www.angstronmaterials.com/shop/ang-stp-k10-graphene-thermal-paste/
 
capt_cope said:
You should try out the graphene paste. I've got a tube from these guys (it was a promo so I'm not sure if they still sell it in "consumer" sizes) and it's worked well for me. I haven't had enough time to test all the variables, but I did see a massive drop when I switched from my standard as5. I'd love to know if that was just improper mounting the first time, or if this stuff has merit.

https://www.angstronmaterials.com/shop/ang-stp-k10-graphene-thermal-paste/
Click to expand...
Well, if it was for sale in the US I would buy it, but it is not.
 
capt_cope said:
You should try out the graphene paste. I've got a tube from these guys (it was a promo so I'm not sure if they still sell it in "consumer" sizes) and it's worked well for me. I haven't had enough time to test all the variables, but I did see a massive drop when I switched from my standard as5. I'd love to know if that was just improper mounting the first time, or if this stuff has merit.

https://www.angstronmaterials.com/shop/ang-stp-k10-graphene-thermal-paste/
Click to expand...
Whats crazy is that it has been available since 2013 yet there isnt a single review.
It must be crap.
 
capt_cope said:
doh.

There *was* a distributor here in the midwest. Didn't realize it isn't for sale anymore.

How much do you need for a test? I've easily got enough left for one or two more applications.
Click to expand...
I need enough of it for it to be for sale in NA.
 
I'd be curious to see how high end thermal pads compare to typical paste TIMs. Fujipoly Thermal pads are available up to 17W/mK. Compared to silicone based grease TIMs at 5W/mK, going by these numbers, the thermal pad should outperform the silicone TIMs.

http://www.frozencpu.com/products/1...rmal_Conductivity_170_WmK.html?tl=g8c487s1797

I just made a thread a few minutes ago asking if anyone had experience with this, but I suppose if it's added to the round up we'll get a real answer.
 
Ej24 said:
I'd be curious to see how high end thermal pads compare to typical paste TIMs. Fujipoly Thermal pads are available up to 17W/mK. Compared to silicone based grease TIMs at 5W/mK, going by these numbers, the thermal pad should outperform the silicone TIMs.

http://www.frozencpu.com/products/1...rmal_Conductivity_170_WmK.html?tl=g8c487s1797

I just made a thread a few minutes ago asking if anyone had experience with this, but I suppose if it's added to the round up we'll get a real answer.
Click to expand...
All of those I have been exposed to in my experience have basically been a joke. I have no interest in covering those. That said, they have been out for a couple years, and if they were as good as they say they are, I am sure we would have some evidence already.
 
I see you have coolabs liquid copper, but i dont see the Ultra Liquid Pro, which IMO i have found to be one of the best for delidding, i get like 20 degree deltas on load
 
thesmokingman said:
FYI, baby wipes clean CLU like a babies butt, no joke.

Btw, on the topic of ICD, have you seen the massive fail thread at TPU? It's glorious in its degree of ICD management fail.



Give baby wipes a try with CLU. :eek:
Click to expand...

:ROFLMAO:

alright alright

I have to remove my air cooler when my AiO arrives

gonna give it a try
 
thesmokingman said:
FYI, baby wipes clean CLU like a babies butt, no joke.

Btw, on the topic of ICD, have you seen the massive fail thread at TPU? It's glorious in its degree of ICD management fail.



Give baby wipes a try with CLU. :eek:
Click to expand...
I would be very hesitant of that given those are designed to leave all sorts of things behind to control diaper rash and moisten the skin. I will stick with cleaning with an highly aromatic solvent.
 
You must log in or register to reply here.
Back
Top