thermal paste alternative?

[krader]

in a previous LTT video they tested various items to make DIY thermal paste. 

 I believe they overlooked one key product that might have gotten them better results. Anti-seize lubricant, its usual application is for bolts and other fasteners so they don't......well......seize. they come in various "grades" it looks like thermal paste and has similar qualities. i think they should test it for the fun of it.

 

[G3R489]

I have a spare CPU and mobo laying around so I could test it for you at some point.
Can’t really justify getting some at the moment, but once payday comes I’m happy to give it a go for you.

 

However how ‘viscous’ is it, as if it’s super runny or similar to a liquid it could short things, and I’d rather not do that if I can avoid it.

[Bombastinator]

23 minutes ago, krader said:

in a previous LTT video they tested various items to make DIY thermal paste. 

 I believe they overlooked one key product that might have gotten them better results. Anti-seize lubricant, its usual application is for bolts and other fasteners so they don't......well......seize. they come in various "grades" it looks like thermal paste and has similar qualities. i think they should test it for the fun of it.

 

There is another video where an engineer attempts to make thermal pastes using various materials.  The best he was able to come up with was a thermal epoxy

 

[LogicalDrm]

-> Moved to Air Cooling

[thermalgoop]

Weld the cooler to the IHS. 

[freeagent]

You don't want to weld it, the heat will cause too much distortion. It would be tough to get an arc in there anyways :D

[Guest]

I always wondered if copper anti seize paste would work well, you know the stuff you use on brakes.

 

In fact, I have an old ryzen and some stock coolers lying around.... I might try that out this weekend.

 

Edit: Maybe not, I just googled it

[krader]

On 10/21/2020 at 6:53 PM, G3R489 said:

I have a spare CPU and mobo laying around so I could test it for you at some point.
Can’t really justify getting some at the moment, but once payday comes I’m happy to give it a go for you.

 

However how ‘viscous’ is it, as if it’s super runny or similar to a liquid it could short things, and I’d rather not do that if I can avoid it.

it has the consistency of thermal paste not runny. they make a non conductive version among others. id rather you not take the risk of ruining your parts just to satisfy my curiosity.

[krader]

12 hours ago, Shabba said:

I always wondered if copper anti seize paste would work well, you know the stuff you use on brakes.

 

In fact, I have an old ryzen and some stock coolers lying around.... I might try that out this weekend.

 

Edit: Maybe not, I just googled it

what did google say?

[CommanderAlex]

12 hours ago, Shabba said:

I always wondered if copper anti seize paste would work well, you know the stuff you use on brakes.

 

In fact, I have an old ryzen and some stock coolers lying around.... I might try that out this weekend.

 

Edit: Maybe not, I just googled it

 

20 minutes ago, krader said:

what did google say?

I would imagine you have the risk of shorting something out if it drips due to the fact that copper is a conductor of electricity. 

[thermalgoop]

13 hours ago, freeagent said:

You don't want to weld it, the heat will cause too much distortion. It would be tough to get an arc in there anyways

Lol it was a joke, but it was inspired by some old overclocking forums I was on in the mid 00s. The rather notorious user (that I seem to remember from quite a few forums) Methylphenidate was discussing the Intel Kentsfield thermals and the soldered IHS, and someone seriously suggested welding either the IHS to the die or water block to IHS for the best thermal transfer. Which, theoretically, it would be a good transfer, but there's obviously some rather gaping practicality concerns to overcome. 
Boiling a 6850QX was also advice given in the same thread. Not something I was about to do with mine...

[Turbo.SG]

IC Cooling makes graphite thermal pads. I use them in my systems and haven't had issues yet. Temps aren't too bad either, *See LTT videos*

 

 

[krader]

5 minutes ago, Turbo.SG said:

IC Cooling makes graphite thermal pads. I use them in my systems and haven't had issues yet. Temps aren't too bad either, *See LTT videos*

 

 

i saw the both but it doesn't answer my question lol. i would never use anit-seize on my cpu but i would love to know if it works just because im curious.

[G3R489]

11 hours ago, krader said:

it has the consistency of thermal paste not runny. they make a non conductive version among others. id rather you not take the risk of ruining your parts just to satisfy my curiosity.

If there’s a non conductive version that could be good then.  
And trust me, I have allot of spare old parts, if I can avoid ruining them I will, but if it’s non conductive and has the consistency of thermal paste, I might still give it a try on an old Athlon or something 

[Bitter]

Most anti-seize (like the one shown) are a paste of aluminum powder and heavy oil/light grease. It gets very very thin and turns into basically oil with higher temps. When just sitting in the bottle oils separate out and puddle on top. It's not a very stable emulsion at all and even less so when exposed to higher temps. It would melt between the CPU and cooler likely leaking a slurry of finely powdered aluminum and oils onto the CPU and socket and probably shorting something out. It's super messy too, a tiny bit smears and makes a huge mess and can stain and remain on surfaces with a high difficulty of removal. I'm not sure the aluminum powder is fine enough or uniform enough for good thermal transfer. Where it might work could be possibly for sub zero cooling, if the powdered metal is small and uniform and base oils are low temp stable, but I doubt that's the case either, the oils probably become too thick and may even freeze and crack. I think anti-seize compounds are best relegated to the work shop where they belong.

[yenic]

 

On 10/23/2020 at 9:34 PM, Turbo.SG said:

IC Cooling makes graphite thermal pads. I use them in my systems and haven't had issues yet. Temps aren't too bad either, *See LTT videos*

 

 

I've used the IC graphite pad for years now, since 2017, love it. I agree with the comments here, that anti-seize is probably not formulated to hold up over long periods of time. It'll either dry up or be chemically affected in a negative way by the heat.

 

I used and preferred Arctic MX4 before moving to the thermal pads, because it's one of few that was or has been on the market long enough with many results that I could verify that it didn't need reapplying after 8+ years. I'd rather have that benefit than 1-5C temp improvements. Which is why I use a thermal pad today..

[Bombastinator]

3 minutes ago, yenic said:

 

I've used the IC graphite pad for years now, since 2017, love it. I agree with the comments here, that anti-seize is probably not formulated to hold up over long periods of time. It'll either dry up or be chemically affected in a negative way by the heat.

 

I used and preferred Arctic MX4 before moving to the thermal pads, because it's one of few that was or has been on the market long enough with many results that I could verify that it didn't need reapplying after 8+ years. I'd rather have that benefit than 1-5C temp improvements. Which is why I use a thermal pad today..

I’ve had really good luck with diamond fill stuff. Unfortunately I don’t remember the brand.  Really long life is a good argument for mx4

[yenic]

16 hours ago, Bombastinator said:

I’ve had really good luck with diamond fill stuff. Unfortunately I don’t remember the brand.  Really long life is a good argument for mx4

I think it's just Innovation Cooling Diamond? Same source as my thermal pad. I never used IC Diamond but it's definitely one of the best (or was, I don't keep up on this too closely), I did read it's harder to spread than other options. That wouldn't bother me though.

 

One reason I was happy to move to the 40x40mm IC Graphite pad was it covers a Ryzen heatspreader completely. I've always been paranoid I put too much or too little, and that I definitely got paste over the dies themselves. Before these graphite pads, I'd research the die layout to make sure the dot or line would cover that portion for sure. Not sure it matters much in reality but it's a nice perk to just have perfect coverage.

The IC Graphite pads are electrically conductive but it doesn't make any sense to worry about that when there's no squish going on. I think my 40x40 pad does have slight overhang on my 2700X IHS but not enough to even come remotely close to touching anything.

 

I'd use MX4 again, I never did find any other brand that had as many users that vouched they took off their heatsink and it was still viscous after 8+ years. I think "Arctic" actually guarantees it for that long too on the package, 8 years (yup just checked their site). I also liked that it wasn't electrically conductive so even if less-is-more with paste, you can put extra if you feel the urge. It seemed to perform as well as "Arctic Silver" which is conductive, and I've pulled off heatsinks to find Arctic Silver had dried up.

 

I really can't see myself going back from these thermal pads for CPUs. If I redid a GPU's paste (or anything that isn't a CPU), I would probably use MX4 there. Just for ease of fit in small/oddly shaped spaces, and so I don't have to think about it again. I don't overclock anymore and losing even 5C, which would be overstating the difference for sure, means nothing to me.

[Bombastinator]

44 minutes ago, yenic said:

I think it's just Innovation Cooling Diamond? Same source as my thermal pad. I never used IC Diamond but it's definitely one of the best (or was, I don't keep up on this too closely), I did read it's harder to spread than other options. That wouldn't bother me though.

 

One reason I was happy to move to the 40x40mm IC Graphite pad was it covers a Ryzen heatspreader completely. I've always been paranoid I put too much or too little, and that I definitely got paste over the dies themselves. Before these graphite pads, I'd research the die layout to make sure the dot or line would cover that portion for sure. Not sure it matters much in reality but it's a nice perk to just have perfect coverage.

The IC Graphite pads are electrically conductive but it doesn't make any sense to worry about that when there's no squish going on. I think my 40x40 pad does have slight overhang on my 2700X IHS but not enough to even come remotely close to touching anything.

 

I'd use MX4 again, I never did find any other brand that had as many users that vouched they took off their heatsink and it was still viscous after 8+ years. I think "Arctic" actually guarantees it for that long too on the package, 8 years (yup just checked their site). I also liked that it wasn't electrically conductive so even if less-is-more with paste, you can put extra if you feel the urge. It seemed to perform as well as "Arctic Silver" which is conductive, and I've pulled off heatsinks to find Arctic Silver had dried up.

 

I really can't see myself going back from these thermal pads for CPUs. If I redid a GPU's paste (or anything that isn't a CPU), I would probably use MX4 there. Just for ease of fit in small/oddly shaped spaces, and so I don't have to think about it again. I don't overclock anymore and losing even 5C, which would be overstating the difference for sure, means nothing to me.

I don’t know the brand.  I never bothered to spread it.  I used the pea method. I know when I bought it there were a bunch of brands of industrial diamond filled paste.  I went for the smallest size and largest percentage I could get.