Measure thermal pastes and pads: ASTM D5470-17 and the Nanotest TIMA5 – Correctly determine bulk and effective thermal conductivity

1 - Introduction and efective thermal resistance 2 - Calculation of the interface resistance 3 - Thermal bulk resistance and bulk conductivity 4 - Equipment, datasheets and conclusion

I will use today’s article as a reference for the devices and methodology used in the upcoming databases for thermal interface materials (TIM) and explain the entire measurement process in transparent detail based on questions from readers and objections from the forum. Of course, there will be no getting around certain mathematical formulas, but for anyone who finds this too complex, I have also created simple diagrams and schematic graphics to illustrate everything better. So don’t be afraid to read on, because it’s worth it!

Even though these methods and the calculation principles are actually trivial, I unfortunately had to realize that there are still gaps in the general approach to the problem of correctly determining the value of thermal resistances and thermal conductivities. I was also surprised that even readers who have been writing technically sound articles for years find it so difficult to deal with the low (and therefore real) values for thermal conductivity that I have determined. But it also shows that the marketing of the gaming and accessories industry works better than you might like. Of course, that’s no big deal, because you can explain everything. And that’s exactly what I’m going to do today, because we need to make a very good distinction between practical values and what is used in theory as a marketing statement.

I generally measure the thermal pastes for my articles and the database in accordance with ASTM D5470-17 and I also try to reduce most negative influences in advance. This is exactly why I work with an initial layer of 500 µm after the respective calibration, which I first heat slowly to 120 °C without much pressure, then cool down to 20 °C to finally heat it to the constant 60 °C of my measurements. Only then do I measure the thermal resistance or thermal conductivity under identical laboratory conditions at a constant 60 °C average paste temperature and from 400 µm downwards in steps of 25 µm. This is done in a standardized way, whereby all interfering factors (such as die distortions or non-coplanar contact surfaces) can be excluded. Controlled surface conditions, unidirectional heat flow conditions, parallel contact surfaces and precisely known clamping forces are guaranteed.

I use the TIMA5 from Nanotest, a compact all-in-one benchtop device that combines the measurement setup and the required PC in one device. It is a self-sufficient and, above all, automated measurement setup that I can also run in parallel to other tasks in the background. After all, who wants to sit and watch for 6 hours or more? A test series like this is virtually impossible to carry out manually. All data is saved directly to the NAS via the network. The device is recalibrated before each measurement (sensors for pressure and BLT). But what is actually behind this? We now need to find out.

Thermal resistance, interface resistance andeffective thermal resistance

Let’s start with the general description A so-called thermal resistance describes the resistance that a material offers to heat transfer. It is based on the thickness d of the material, its thermal conductivity λ andthe area A over which the heat flows:Here, the normal thermal resistance ignores the contact points between different materials, which is idealized and does not occur at all in practice. This is theory and is of no use to me in this form for the time being, because a further resistance comes into play. The so-called interface resistance arises because heat is transferred less effectively at the point of contact between two materials than in the material itself. This can be caused by uneven surfaces or incomplete contact at the interface (e.g. microscopic irregularities). For example, in PC systems where heat is transferred from a processor to a heat sink, the interface resistance can significantly increase the overall thermal resistance. We also have the factor of different media when the heat transfers from a solid to a liquid medium and vice versa

Let us now turn to the conceptual explanation of what I can measure. The effective thermal resistance also takes into account the interface resistance _Rinterface at the interfaces between the materials, as just explained. This interface resistance occurs at the point of contact between two materials and arises due to the thermal conductivity of the materials. The effective thermal resistance is therefore determined by the sum of the normal thermal resistance of the material and the interface resistance:The interface resistance can make up a significant part of the total resistance, particularly with very thin layers (BLT, bondline thickness), especially when two materials with different thermal properties come into contact. I cannot measure this important value or read it out of the total value, but I can calculate it later.

The measurement according to ASTM D5470-17 in practice

The ASTM D5470-17 method with six temperature sensors makes it possible to precisely determine the effective thermal resistance of a TIM by measuring the heat flow and temperature distribution along the metal blocks. The interface resistance at the interfaces is also taken into account to ensure the most accurate measurement of thermal resistance. Let us now take a brief look at the schematic and practical setup of the TIMA5 and the measurement of the TIM between two test heads with three sensors each:

In this method, the effective thermal resistance of a TIM is determined between two solid surfaces that are loaded by heat flow. In the configuration with six temperature sensors, the temperature is measured at various points along the heat column in order to calculate the temperature gradient and thus the thermal resistance. A sample of the TIM is placed between two metal blocks, which are heated by a heating element at the top and cooled by a laboratory water cooler at the bottom. The heat flows through the TIM from one hot side to the cold side. Three sensors are typically located on the hot side of the TIM and three on the cold side.

The temperature distribution in the metal block follows a linear temperature curve, as the blocks are made of a homogeneous material (copper) with known thermal conductivity and have all been measured beforehand and provided with a calibration file. The temperature gradient in the metal block is now used to determine the heat flow q. This heat flow q is calculated using the measured temperature gradient in the metal block and the known thermal conductivity λ of the metal, where ΔTis thetemperature difference and dBlock is the thickness of the metal block:The temperatures directly at the top and bottom surfaces of the TIM (i.e. at the interfaces with the metal block) are extrapolated from the measurements. This provides the temperature at the interface, which is used to calculate the temperature drop in the TIM. The effective thermal resistance _Rth of the TIM is simply calculated by the temperature drop _ΔTTIMover the thickness of the TIM _dTIM, whereby the heat flow q is known. Here, ΔTTIM is the difference between the extrapolated temperatures on the top and bottom sides of the TIM: I can also use a curve diagram for the effective thermal resistance, which is much more relevant in practice than the bulk value (best 1o of the pastes tested so far).

Finally, the thermal conductivity _λTIM of the TIM can be calculated if the heat flow and the thermal resistance are known. Here, A is the area of the TIM sample:

In a curve diagram for the effective thermal conductivity, which is actually much more relevant in practice than the absolute bulk value, it then looks like this (best 1o of the pastes tested so far). And for those who are now wondering: the thicker the BLT, the lower the influence of _Rinterface, logically:

The interface resistance plays an important role in ASTM D5470-17. In addition to the thermal resistance of the TIM itself, TIMA5 always has an additional thermal resistance at the interfaces between the TIM and the two metal blocks. However, this can be determined separately by carrying out several measurements with different thicknesses of the TIM and extrapolating the results to determine the interface resistance at a bondline thickness (BLT) of 0. This is exactly what you will find out after turning the page…