[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

[Shrek]

BTW Andy you should try SYY-157.

And therein lies my problem... sure I could look at temperatures, but what really counts for longevity is pump-out and that needs thousands of thermal cycles.

I don't mean to downplay all the work on temperatures, but it is a bit like comparing one car against another by top speed when there are so many more criterion, many much more important. That is why I keep returning to longevity (leaving cost aside)

This is where I am wondering if a future generation of TIMs will include surfactants that will have the surface tension refill the gap between chip and heatsink.

 

[freeagent]

You would know when you saw it that it won’t suffer the effects you speak of

 

[lexluthermiester]

BGA laptop?

Both are BGA...

Because BGA Was bad according to rooter

Those were his tests. Those results are relative to the methodology used to do the testing which I suspect has a few flaws.

 

[Shrek]

You would know when you saw it that it won’t suffer the effects you speak of

I would almost disagree with you, but I recall the move from conventional motor oil to synthetic and how much more the synthetic would stick; that is why I have been bringing up the idea of surfactants.

OK, now I agree with the earlier comment that I am overthinking this all.

Longevity tests (including pump-out) are simply outside of the hobbyist ability and we need to leave that to the likes of industry.

 

[tabascosauz]

Pretty sure the heatsink fit is everything here. Some of these dellienware tripod heatsinks are worse than atrocious.

I do agree with applying to both surfaces. For MX-5 and Kryonaut Extreme I just do a normal thin layer on the heatsink and a thicker layer on the GPU.

That's a good point. You'd think it obvious but apparently not for my smooth brain

When I repasted my XPS 13 9370 I found that Dell was using this hard slab of "thermal paste" that was at least 1mm thick. Had to scrape it off the copper contact plate with a metal spudger, it was that tough even against 99%IPA. I watched some Conductonaut tutorials for the 9370 on Youtube and it looked like they just applied the usual thin layer on both die+heatsink so I just gave it a liberal dose of MX-4 thinking it was enough. Smooth brain moment.

Put it together and it's been getting even worse thermals than stock ever since. Now that I have time I've been considering experimenting with Conductonaut, but now that you mention it, I think I'll try putting a layer on both the heatsink and the die. Problem is, I worry that MX-4 is too runny for this, but it's the only paste I have in sufficient quantities for this.

And the fact that it's a pain to open up because I cut 8 of my fingers at once on the knife-sharp inner edge of the aluminium bottom cover last time not using gloves. But I reckon I should give it another shot since it'll probably be replaced by an iPad anyways in the near future now that studies are finally over.

If it's too runny I suppose I could buy a tube of NT-H2.

 

[Shrek]

Want a really crazy idea? how about using a ferro-fluid with a magnet mounted at the center of the heatsink

Ferrofluid - Wikipedia

Now the fluid will return after being pumped out...

I found that Dell was using this hard slab of "thermal paste" that was at least 1mm thick.
Put it together and it's been getting even worse thermals than stock ever since.

Amazing how maybe our fancy stuff can't outdo what we thought was bad.

 

[lexluthermiester]

OK, now I agree with the earlier comment that I am overthinking this all.

To be fair, at least you're thinking things over. There are those who spout out rubbish without thinking things through. So no worries.

 

[Shrek]

At the risk of repetition, looking again at my measurements

Thermal Interface Material (TIM)	Temperature
None	84°C
SuperLube thermal compound	72°C
Petroleum Jelly	71°C
Diaper cream (40%)	68°C
GD900	68°C

Taken at face value, I would be using diaper cream or petroleum jelly over SuperLube thermal compound; for me that emphasizes the dangers of just relying on temperature performance as I have come to the absolutely ridiculous conclusion that diaper cream is a good TIM

This is the reason why I feel it is so very important to include diaper cream.

 

[Borc]

If you don't like what I posted, buy your own samples of TIMs, do your own tests and post your own results.

I'm doing all of this, don't worry. I'm just curious what do you mean with impressive, is it too much to ask for? You would need to compare MX5 with several of the top performer to really know. Noctua NT-H1 has been released in 2007, it's old. If MX5 is on par with this paste it would imply MX5 is down several degrees compared to the best. Try out SYY-157 if you can.

 

[lexluthermiester]

I'm doing all of this, don't worry.

Make sure you take pictures and screenshots.

I'm just curious what do you mean with impressive, is it too much to ask for?

Given the attitude you displayed? Hell yes.

You would need to compare MX5 with several of the top performer to really know.

No. I do as I see fit and compare to what I had on hand.

Noctua NT-H1 has been released in 2007, it's old.

I actually ordered 3.5g NT-H2 but Amazon canceled the order and I wasn't going to spend more on what ultimately was a personal experiment. Additionally, AGE does NOT define usefulness. Arctic Silver 5 is STILL a competitive TIM after decades. Let that sink in.

If MX5 is on par with this paste it would imply MX5 is down several degrees compared to the best.

Yes, several degrees, as in less than 5C. Given testing found elsewhere on the net, MX-5 is less than 4C away from the top performers and yet costs a fraction of those "top tier" brands.

The reasons for doing this testing are defined in the edit at the bottom of the results post(#520). I didn't do it to be competitive, nor am I going to explain myself to anyone giving a less than friendly/respectful attitude.

Try out SYY-157 if you can.

No Thanks. I'm complete satisfied with what I've got..

 

[ThrashZone]

At the risk of repetition, looking again at my measurements

Thermal Interface Material (TIM)	Temperature
None	84°C
SuperLube thermal compound	72°C
Petroleum Jelly	71°C
Diaper cream (40%)	68°C
GD900	68°C

Taken at face value, I would be using diaper cream or petroleum jelly over SuperLube thermal compound; for me that emphasizes the dangers of just relying on temperature performance as I have come to the absolutely ridiculous conclusion that diaper cream is a good TIM

This is the reason why I feel it is so very important to include diaper cream.

Hi,
With diaper cream around maybe add vagisil too

 

[Shrek]

And here was me thinking that maybe I had a valid point with 40% Zinc Oxide; I should have stuck with Zinc Oxide ointment, although most is only 20% strength.

My point is that all these temperature tests are just a small part of the story if 40% 'ointment' looks so good under such tests, but for some strange reason nobody else seems willing to try.

The joke is on me.

 

[lexluthermiester]

The joke is on me.

Not really. The reason I originally tested the copper grease is because I already had some and then bought different branded tube for giggles. My main concern was this: You are suggesting using a product that is formulated for use with human skin. Copper grease is formulated to be used in automotive and industrial applications, so transferring usage to a computer environment is not a stretch. Transferring a skin cream to use in a PC is more worrisome as the chemistry for the suspension material might react poorly to the temperature ranges involved. Making something for use in machines is very different than making something for use on skin.

It's not a risk I'm willing to take.

 

[Space Lynx]

Not really. The reason I originally tested the copper grease is because I already had some and then bought different branded tube for giggles. My main concern was this: You are suggesting using a product that is formulated for use with human skin. Copper grease is formulated to be used in automotive and industrial applications, so transferring usage to a computer environment is not a stretch. Transferring a skin cream to use in a PC is more worrisome as the chemistry for the suspension material might react poorly to the temperature ranges involved. Making something for use in machines is very different than making something for use on skin.

It's not a risk I'm willing to take.

when are you expecting your tests to be posted?

 

[lexluthermiester]

when are you expecting your tests to be posted?

What? I did. See post #520. Did you really miss them?

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

Oh I'll pass on that. That's specifically for TFX. (I got the best results on my 3090 FE (most stable core temp deltas) applying it like that. SYY-157 still needs full coverage but is easier to get on. Anyone here use SYY-157 yet?

www.techpowerup.com

 

[Space Lynx]

What? I did. See post #520. Did you really miss them?

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

Oh I'll pass on that. That's specifically for TFX. (I got the best results on my 3090 FE (most stable core temp deltas) applying it like that. SYY-157 still needs full coverage but is easier to get on. Anyone here use SYY-157 yet?

www.techpowerup.com

I did. I can't keep up with this thread. lol

 

[Falkentyne]

What? I did. See post #520. Did you really miss them?

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

Oh I'll pass on that. That's specifically for TFX. (I got the best results on my 3090 FE (most stable core temp deltas) applying it like that. SYY-157 still needs full coverage but is easier to get on. Anyone here use SYY-157 yet?

www.techpowerup.com

I'm still curious how this material will perform. The Kopr-Kote was tested, but this one seems interesting to see if it's any better. Still hoping I can test it next month if my living situation improves.

Amazon.com: Jet-Lube SS-30 - Pure Copper | High Temperature | Anti-Seize | Military Grade | Environmentally Preferred | Thread Compound | 1/4 Lb.: Industrial & Scientific

Amazon.com: Jet-Lube SS-30 - Pure Copper | High Temperature | Anti-Seize | Military Grade | Environmentally Preferred | Thread Compound | 1/4 Lb.: Industrial & Scientific

www.amazon.com

 

[crispysilicon]

And here was me thinking that maybe I had a valid point with 40% Zinc Oxide; I should have stuck with Zinc Oxide ointment, although most is only 20% strength.

My point is that all these temperature tests are just a small part of the story if 40% 'ointment' looks so good under such tests, but for some strange reason nobody else seems willing to try.

The joke is on me.

Try https://www.permatex.com/products/l...i-seize/permatex-copper-anti-seize-lubricant/

 

[Shrek]

It's not a risk I'm willing to take.

I respect your opinion, even if I don't understand it.

Try https://www.permatex.com/products/l...i-seize/permatex-copper-anti-seize-lubricant/

The point I was/am trying to make is that temperature performance is not very useful; 40% ointment was all I needed to make that point.

 

[Falkentyne]

Only suggesting a few minutes run and then at less than 100C; but I respect your opinion, even if I don't understand it.

You need to do these tests yourself, Andy, not ask other people to risk their hardware for you. It's really not nice to expect someone to do that, especially in this current market.
I'm already willing to buy that 30% copper anti-seize thing I posted above, when my income situation allows me to, and that's already putting hardware at risk using stuff like that.
I almost *destroyed* my (almost worthless now) backup video card trying to do this crap for someone who suggested a certain paste would be a good idea after the material hardened completely and the card reached 95C in 5 seconds...

 

[Shrek]

I ALREADY DID! Now I'm getting annoyed.

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

Doesn't that happen with all compounds eventually? Depends on the compound. I thought drying/caking was bad. It happened to me with AS5. Mine killed a gpu, but back then they only held it down with spring loaded pins. If it was a video card that didn't have a fan or had a crap fan, that's...

www.techpowerup.com

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

BTW Andy you should try SYY-157. And therein lies my problem... sure I could look at temperatures, but what really counts for longevity is pump-out and that needs thousands of thermal cycles. I don't mean to downplay all the work on temperatures, but it is a bit like comparing one car against...

www.techpowerup.com

 

[lexluthermiester]

I did. I can't keep up with this thread. lol

No worries! I know how you feel!

I respect your opinion, even if I don't understand it.

It's a risk factor thing. I use the system that was utilized for testing. When older, it's still a usable system I care about. Not eager damage it. Skin creams are just too risky for my comfort level.

 

[Shrek]

That's fine, I've already done the testing and was amazed how it was as good as GD900. I was just wondering if someone wanted to check the results and while I feel there is no short term risk I did say I respected the decision not to try.

That said, I only tried it for a few minutes to get the measurement and then moved to the next trial; I suspect it is not a good long term solution (the whole point I am trying to make concerning longevity).

 

[lexluthermiester]

I ALREADY DID! Now I'm getting annoyed.

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

Doesn't that happen with all compounds eventually? Depends on the compound. I thought drying/caking was bad. It happened to me with AS5. Mine killed a gpu, but back then they only held it down with spring loaded pins. If it was a video card that didn't have a fan or had a crap fan, that's...

www.techpowerup.com

[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

BTW Andy you should try SYY-157. And therein lies my problem... sure I could look at temperatures, but what really counts for longevity is pump-out and that needs thousands of thermal cycles. I don't mean to downplay all the work on temperatures, but it is a bit like comparing one car against...

www.techpowerup.com

While I feel you, this thread is 23 pages long and we can't expect everyone to have read through the whole thing... Heck, @lynx29 missed my huge test results spread entirely and he was looking forward to it! LOL!

That's fine, I've already done the testing and was amazed it was as good as GD900

My point is made; I was just wondering if someone wanted to check the results.

You have given none of us any reason not to trust your results. The reason I went through the effort of posting screenshots is for due diligence but also because there are people in this forum who don't like me and have a habit of nit-picking what I say so I wanted to shut them up before they complained... Taking screen shots and pictures is a good rule of thumb anyway, but neither of us are conducting professional reviews, this is all out of fun. So it's all good mate, no worries!

 

[Shrek]

While I feel you, this thread is 23 pages long and we can't expect everyone to have read through the whole thing... Heck, @lynx29 missed my huge test results spread entirely and he was looking forward to it! LOL!


You have given none of us any reason not to trust your results. The reason I went through the effort of posting screenshots is for due diligence but also because there are people in this forum who don't like me and have a habit of nit-picking what I say so I wanted to shut them up before they complained... Taking screen shots and pictures is a good rule of thumb anyway, but neither of us are conducting professional reviews, this is all out of fun. So it's all good mate, no worries!

You are right, it has been a long and winding road; I should not expect everyone to have read all 23 pages.

Trust no one... haven't you been watching X-files ; anyhow, you know how science goes, results need confirmation.

I actually find it utterly ridiculous that ointment should be able to run with the pack... (short term)


Someone might want to try dielectric grease; it claims high thermal conductivity (this I don't believe) but it is good to 400°F
Technical_Data_Sheet_Silicone_Dielectric_Vacuum_Grease.pdf (super-lube.com)
0.16 W/K m

I already tried petroleum jelly and it wasn't that great even if it beat SuperLube thermal compound, and yes I understand why one might not want to risk trying this.