Real laboratory tests of thermal pastes on igor’sLAB – Part 2 – First comparative measurement with 2 expensive pastes

1 - Thermal resistance instead of thermal conductivity 2 - Practical tests and summary

The advantage of many measuring points – derivation of practical values

Apart from the fact that I also have the temperatures of the heater and the water, which are of no use to us because they always remain constant, I have my measurement setup with temperature sensors 1 to 6 (see diagram below). With these values you can now also make some very nice considerations. Let’s go!

Let us first take the values of T3 and T4, which show us the two temperatures at the respective contact surfaces between which the paste is located. These curves are no longer completely linear, because the interface resistance also changes a little. And we do not calculate with 6 points, but only with 2 absolute values for the temperature difference instead of a gradient as with_TTim, whereby the sample temperature remains constant. And what is the point of all this? The behavior is similar to that of a graphics card, which has to manage without an IHS and where the delta is usually measured between the substrate and the water temperature. I will also include this value in my paste tests and the online database.

Now I compare T1 of the reference with T1 of the gaming paste A. While the heater remains constant, we already have sufficient thermal resistance in the copper reference block to simulate the CPU temperature and its differences with different pastes in comparison to the reference and in relation to the layer thickness of the paste. It is precisely this variable evaluation that no test on a CPU can offer, because it is always individually different and therefore not really reproducible. Here in the TIMA5 test, however, it is.

What also seems interesting is the contact resistance, in our case the interface resistance. Here you can see how well the surface of the paste “clings” to the contact surfaces (IHS, heatsink). These values are also easy to compare and meaningful, as they are always the same calibrated reference blocks. Coarser degrees of grinding or a less favorable microstructure can also be a negative factor, which then influences the effective thermal resistance and thus also the conductivity, as can too low temperatures and too high a viscosity.

Summary and conclusion

The first thing I noticed was that all the data on thermal conductivity is actually superfluous. If a paste spreads more easily and more thinly, then it can certainly perform at the level of a nominally better paste, which is perhaps significantly more viscous and has a poorer interface resistance. I did the counter-test on a water-cooled Intel Core i9-13900K with these two pastes, where in the end there was only one Kelvin difference in CPU temperature between the sample and the Gaming Paste A. We have known for some time that the differences between the pastes are not that great in reality, as long as they do not bleed or dry out and crack.

In addition to these six tests and the evaluation, there will soon be an online database, which of course also contains further information (manufacturer’s data sheet), plus our own photos, the TIMA5 documents and which allows sorting or searching for various factors in the website interface. In addition, there are of course the material tests with the analysis of the paste content and some high-resolution microscopy for the gallery. I can also carry out ageing tests for certain pastes and suspicious moments, but as always this is purely a question of time. But that will probably be a case-by-case decision, depending on the case.

So, I’m going to take a lot of cloths, some delicious isopropyl alcohol and some time to dispose of today’s sour cream from the TIMA5. After all, I measure from 400 µm downwards and that’s why I have to put a lot of paste on it first. Of course, I clean the reference bodies every time, but there is also a lot of spillage. Of course, this also needs to be disposed of at the end of each session. And cleaning the windows is a point of honor. Tomorrow, Wednesday, there will be a case lunch and on Thursday and Friday I will reveal the secret of the reference paste and the Gaming Paste A. After that, depending on the availability of pastes, there will be lots of individual tests, so this will certainly be a long affair.

If you have any suggestions or ideas for the test content or the database content, please feel free to contribute. But what I definitely won’t be doing is testing on a CPU or GPU. I can also emulate this here, but with reliably reproducible values, regardless of whether it’s a cold 10 °C or a Mediterranean 30 °C in the room. That’s what chillers and heaters and all kinds of measurement and control electronics are for. Apart from that, I’m glad that everything is finally running as it should. And a little more transparency in the paste gold market really can’t hurt.

Once again, I refer you to my request for more pastes, because the selection can never be big enough. What’s more, I will soon have a hard-working helper who has already proved his worth as Padman Junior. But don’t worry, everything will definitely stay in the family! And pastes are grateful, I can prepare them nicely without the launch stress. Because next week I’ll be on vacation for a short while, then there will be articles from the team and new articles from me from the canned pastes. So as usual and hopefully without a gap 🙂