The Official Thermal Interface Material thread

[P4-630]

Maybe someone should just weld the cooler to the CPU...

 

[ShrimpBrime]

Neither thermal grizzly nor arctic give out thermal conductivity values for their traditional pastes afaik
I also found that PTM stuff with "only" 8.5W/mK performs better than the pastes that i had with 15+W/mK for example (to be fair difference is quite small, ~2°C but it's there)

We where talking about low end pastes, which do give the details, all of 3.3w/mk....

Since grizzly doesn't give the specs specifically, doesn't mean it's bad paste or has a low w/mk rating. Cause you will tell the difference from 3.3 to 8 and more so 3.3 to 12+.

Silicon - solder - copper - TIM - copper - water - aluminum/copper - air.

My point in another post, thermal interface material is the slow down point.

Lapped cpu and cooler with LM for a long term setup should be most desired in my opinion.

 

[Bagerklestyne]

KryoSheet doesn't work for me, 100 degrees after a few seconds. Looks like there is no contact, it's too thin for my subpar Dell cooler. There is only a 0.2mm version, I can't even choose a thicker version. I tried EC360 Graphite which is even thinner with the same bad result. I think it's not worth the hassle on an uneven laptop cooler or subpar cooler in general, PTM7950 does the job.

That's why many pastes doesn't work great either on my laptop. Some pastes can fill out bigger gaps, lower viscous pastes have no chance. I have read DOWSIL TC-5888 can fill gaps of up to 0.5mm.

Thermal Hero has some Graphite pads listed with a thickness of 0.3mm, maybe I will try this out next year.

This is going to sound like an expensive and stupid idea (I have no clue if it's worth even trying, might have to ask de8auer and see what he thinks) but maybe 2-3 sheets stacked might be viable/effective (albeit pricey)

 

[freeagent]

SYY-157.. real tough to beat.. one of if not the best I have used.. 15.7 w/mk for the jaded ones

And yeah.. I know what Toms says

 

[sLowEnd]

They would use the same standard of testing that would be used for any other tested material out there. Why would it need to be different with thermal paste? Please explain.

That's the thing, there is no consensus on what testing parameters to use, nor do any of the manufacturers publish how they do their testing. Same issue with airflow (CFM) and noise (dba) claims manufacturers make exists with fans.

 

[freeagent]

Well.. I could just as easily say that I do not use my fans as reviewers test them. I actually let them move air. No better way to test TIM yourself than by throwing a few hunnit watts at it and see what it does. Pretty easy.

 

[sLowEnd]

Well.. I could just as easily say that I do not use my fans as reviewers test them. I actually let them move air. No better way to test TIM yourself than by throwing a few hunnit watts at it and see what it does. Pretty easy.

That has nothing to do with how you can't take one brand's claims of ##w/mK and compare it with another brand's claims of ##w/mK with certainty. Thermal conductivity is affected by factors such as bond line thickness and temperature tested at. There is no industry-wide testing standard that these companies adhere to, and nobody publishes testing methodology.

 

[lexluthermiester]

Anyone heard about this stuff? Seems pretty cheap on Amazon, but I want to see if there's any discussion about this stuff anywhere.

8618 - Ultra Thermal Paste, 6 W/(m·K)

8618 silicone-free thermal grease is a heat sink paste with very high thermal conductivity and exceptional wetting properties.

mgchemicals.com

MGChem is a solid and reputable company, so at the very least it will perform as it should and likely be a quality product.

Maybe someone should just weld the cooler to the CPU...

That has been tried... Sort of. Soldered.

 

[freeagent]

That has nothing to do with how you can't take one brand's claims of ##w/mK and compare it with another brand's claims of ##w/mK with certainty. Thermal conductivity is affected by factors such as bond line thickness and temperature tested at. There is no industry-wide testing standard that these companies adhere to, and nobody publishes testing methodology.

Ok.

 

[ShrimpBrime]

That's the thing, there is no consensus on what testing parameters to use, nor do any of the manufacturers publish how they do their testing. Same issue with airflow (CFM) and noise (dba) claims manufacturers make exists with fans.

I don't understand what you're saying. There are various ways to figure out thermal conductivity of anything and everything. Or how would we know what is an insulator or a conductor. I'd hope that most or all manufacturers either conduct an experiment or they calculate based on already known substance thermal conductivity and properties. You can look up all the different thermal properties of different types of ceramics for example.....

Here's a Wiki How. Perhaps even conduct our own home made experiements.

 

[sLowEnd]

I don't understand what you're saying. There are various ways to figure out thermal conductivity of anything and everything. Or how would we know what is an insulator or a conductor. I'd hope that most or all manufacturers either conduct an experiment or they calculate based on already known substance thermal conductivity and properties. You can look up all the different thermal properties of different types of ceramics for example.....

Here's a Wiki How. Perhaps even conduct our own home made experiements.

Thermal Conductivity in w/mK = [[heat input in watts] * [interface thickness in meters]] / [[area in square meters]*[delta T, temperature difference across interface in kelvin]]

You don't know the heat input used for testing. You don't know the interface thickness used for testing. You don't know the area used for testing. You don't know delta T either. All you have from the manufacturers is a conductivity number they hope you will trust at face value. Arctic has dropped out the farce. It's a shame other companies haven't yet done the same.

Similarly with fan noise, we all know that something that makes noise gets quieter the further away it is moved from the receiver. How far away was the microphone used to measure noise for your fan? Chances are very good you have no idea. You are just given a noise number that manufacturers hope you trust at face value. Some companies even just outright lie. Anyone remember SilenX's ridiculous claims of 90cfm @ 1400rpm & 18dba for their IXP-76-18 fans?

 

[lexluthermiester]

You don't know the heat input used for testing. You don't know the interface thickness used for testing. You don't know the area used for testing. You don't know delta T either. All you have from the manufacturers is a conductivity number they hope you will trust at face value.

This! And Arctic knows it...

Arctic has dropped out the farce. It's a shame other companies haven't yet done the same.

Why doesn’t ARCTIC communicate thermal conductivity values?​

ARCTIC made a conscious decision not to specify any values for the thermal conductivity of its thermal paste and thermal pads, because many manufacturers invent, artificially inflate or embellish this value. Thermal paste has a thermal conductivity of 1 to 4 W/mK. Values outside of this range, such as 12.5 W/mK, are at odds with the truth.
Many competitors quote values above 4 W/mK to suggest better performance. This often leads to false expectations and dissatisfied users.
ARCTIC offers its customers innovative thermal interface materials at the best possible price-performance ratio instead of relying on manipulated performance data.

...which is why I really like that company. Truth in advertising goes a LONG way!

It should be noted, that there are some variations and deviations to thermal performance measurements which have some merit. So MGC's claim of 6W/mK is reasonable when taking the variations the company uses into account. The reality is that the actual kinetic thermal performance of their product is something like 3.17W/mK(me using quick and dirty math) without any deviations factored in. While this is only a guess, it likely somewhere near real-world.

So let's not argue this point folks, it's a real a waste of our time and energy(pun intended).

 

[ShrimpBrime]

Thermal Conductivity in w/mK = [[heat input in watts] * [interface thickness in meters]] / [[area in square meters]*[delta T, temperature difference across interface in kelvin]]

You don't know the heat input used for testing. You don't know the interface thickness used for testing. You don't know the area used for testing. You don't know delta T either. All you have from the manufacturers is a conductivity number they hope you will trust at face value. Arctic has dropped out the farce. It's a shame other companies haven't yet done the same.

Similarly with fan noise, we all know that something that makes noise gets quieter the further away it is moved from the receiver. How far away was the microphone used to measure noise for your fan? Chances are very good you have no idea. You are just given a noise number that manufacturers hope you trust at face value. Some companies even just outright lie. Anyone remember SilenX's ridiculous claims of 90cfm @ 1400rpm & 18dba for their IXP-76-18 fans?

Oh god, you want to get real deep with this.... you don't know if the testing is done at sea level or on mount everest with high or low humidity nor the pressure of the fan if used during the testing.

Of course everyone's Delta T will be different. 3.3 watts across a meter of space sounds like a mule pulling an overloaded carriage all be it known to exactly what Solder and Liquid Metal thermal properties are, still MUCH LOWER than copper BTW. We can take a generalized guess that Ceramic pastes don't often exceed, let's say 6w/mk. Then sprinkle some silver dust in it and call it AS5 at 8.9w/mk.

All the numbers look to add up in my head. What do you think of this approach?

Well.. I could just as easily say that I do not use my fans as reviewers test them. I actually let them move air. No better way to test TIM yourself than by throwing a few hunnit watts at it and see what it does. Pretty easy.

This is exactly how you measure thermal conductivity. What is the rate of BTU/hr across the paste between 2 copper surfaces. You only need to know the thickness of the paste. You know the wattage input. You know copper thermal properties. We know the conductivity of water and air. You can calculate exactly how many BTU you NEED to move, that's the easy part. My 13700K pulls 235w at a full load. That's about 802BTU/hr, just a little more than half of my window air conditioning unit I chill loops with.

If we flipped our window A/Cs around in the winter, we could heat homes and cool the earth down!! (off topic parity sentence lol)

 

[sLowEnd]

We can take a generalized guess that Ceramic pastes don't often exceed, let's say 6w/mk. Then sprinkle some silver dust in it and call it AS5 at 8.9w/mk.

All the numbers look to add up in my head. What do you think of this approach?

I hope you are joking.

I'll stop here. I've already said what I wanted to say in my previous posts.

 

[freeagent]

I'll stop here.

No need to get snobby about it. We are cooling CPUs and GPUs. Whether it says 8w/mk or 20.. the only really way to see if it good is to just run it.. I literally do not care what a reviewer finds. My results are always different anyways.

Why do you think I say AS5 is ok when most people are too busy saying that it’s too old fashioned.. I am not a parrot.

 

[lexluthermiester]

Why do you think I say AS5 is ok when most people are too busy saying that it’s too old fashioned.. I am not a parrot.

Have never said that. AS5 is old sure, but it's still a good performer... If you have some or like using it, use it. It's still good stuff.

 

[P4-630]

AS5 is old sure, but it's still a good performer...

It is.. Also can do many years without re-pasting. I know from self experience.

 

[lexluthermiester]

It is.. Also can do many years without re-pasting. I know from self experience.

Me too! AS5 is great, which is why it's still around. It's not the absolute best performer any more(it was for a long time), but it's still very good and long-lasting.

 

[ShrimpBrime]

I hope you are joking.

I'll stop here. I've already said what I wanted to say in my previous posts.

Right you said an opinion without facts.
You have no facts about manufacturer testing.

I gave facts how you would calculate heat exchange. With links. That contain data.

That's what I was trying to get from you. But was provided a single manufacturers statement which meant absolutely nothing to me at all.

 

[nomdeplume]

[sat in post editor since 1 minute after question was published]

Given the current state of general disrest. I'm abdicating on the same level of answer I hope others see fit to.

 

[mechtech]

Hmmmm solder heatsink to cpu. Never have to repaste. I like it. Just have to find a type of solder that will stick.

SYY-157.. real tough to beat.. one of if not the best I have used.. 15.7 w/mk for the jaded ones

And yeah.. I know what Toms says

Usually inexpensive as well. Never tried it. Finally, finished using the last of my mx4, and got a new tube of mx6 (first one)................maybe I will repaste my CPU over the holidays and see how it does............seems kind of a waste since paste on CPU is only 1 year old.

 

[Robits]

Are there any fake PTM 7950 or Upsiren PCM-1 on Aliexpress? I think I bought a fake PCM-1... https://aliexpress.com/item/1005006011357826.html

 

[Ferrum Master]

Are there any fake PTM 7950 or Upsiren PCM-1 on Aliexpress? I think I bought a fake PCM-1... https://aliexpress.com/item/1005006011357826.html

There are no PTM 7950 or PTM 7958 on aliexpress for sure. They have their own reverse engineered compound ant PCM-1 is one of them imho, they come in two flavors also, one is a bit more performant. But they did the job pretty good this time, their knock off PTM 7950 looses only few C versus the original Honeywell.

 

[Robits]

There are no PTM 7950 or PTM 7958 on aliexpress for sure. They have their own reverse engineered compound ant PCM-1 is one of them imho, they come in two flavors also, one is a bit more performant. But they did the job pretty good this time, their knock off PTM 7950 looses only few C versus the original Honeywell.

Damn, I tought PCM-1 had the same performance as HoneyWell 7950 from watching this video, guess that dude is a paid shill:

 

[Ferrum Master]

Damn, I tought PCM-1 had the same performance as HoneyWell 7950 from watching this video, guess that dude is a paid shill:

It is actually pretty hard to get genuine PTM7958 for most people, but the popular makers are slowly hopping in the train and offering it too. Just wait.

But as I said, the difference ain't that bad, use any you have at hand.