Comparison
General Information
Manufacturer Specifications
Notes and Recommendations
Measurements
Minimum Possible Layer Thickness
This is exactly why I wanted to find out how far one can go with a bit of pressure and how much a putty can still be compressed below 250 µm. Here, I use the usual 9N pro cm², which is more than sufficient and exceeds the pressure typically achieved by, for example, a GPU cooler.
Thermal Resistances Rth
Let's start with the most important aspect, the thermal resistance Rth. The key property of Rth is its linear correlation with layer thickness, while thermal conductivity follows a completely different curve and remains far from linear. For memory, we are interested in layer thicknesses (BLT) of 1000 µm and below; for voltage regulators (VRM), they are usually between 1500 µm and 3000 µm, depending on the cooler design.
I have again illustrated the relevant layer thicknesses from 250 to 3000 µm as a bar chart for Rth for comparison.
Effective Thermal Conductivity and Cooling Simulation
If Rth is already available, one wouldn't actually need λeff, the effective thermal conductivity. We can also observe how the values change across the BLT, although one cannot expect a linear curve due to the included area and BLT.
I have again illustrated the relevant layer thicknesses from 250 to 3000 µm as a bar chart for λeff for comparison.
Memory Simulation (VRAM)
First, we take the two temperatures at the two contact surfaces between which the paste is located and examine the difference (Delta). These curves are no longer entirely linear, as the interface resistance also changes. We now calculate using only two absolute values for the temperature difference instead of a gradient as with TTim, where the sample temperature remains constant at 60 °C. Why do this? The behavior is similar to that of the memory of a graphics card, which must function without an IHS and where the Delta between the substrate and the water or cooler temperature or air is measured. This can be projected quite well, as I test the temperature difference at the two surfaces between which the paste is located.
Voltage Regulator Simulation (VRM)
Now, I compare the surface temperatures of the tested products. When normalizing the values for the TIMA5 heater, we already have sufficient thermal resistance in the copper reference block to simulate the package of a voltage regulator chip (VRM). Depending on the thermal resistance of the putty, the temperature will either increase or decrease, much like what occurs on the VRM of a graphics card or a motherboard.
