The effective thermal conductivity
I have written time and again that almost nothing can be recognized or compared on the basis of thermal conductivity. Anyone who believes that thermal conductivity is a constant value is very much mistaken. But I have already explained this at length in the basic article. If you have Rth, you don’t need λeff, i.e. the effective thermal conductivity, at all. And the pure specification for the idealized bulk value is so far removed from reality that you always have to wonder about these figures. The stated 17 W/(m-K) is guaranteed to have been produced in the PR department and not in the laboratory by a real measurement. Yes, it’s still better than what some “high-performance pastes” really offer, but it’s not enough for the top. Good midfield, but no more.
Of course, the whole thing is also shown again as a bar chart for the four most important layer thicknesses:
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 on page 2). With these values you can now also make some very nice considerations.
GPU emulation
Let’s 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, as the interface resistance also changes slightly. And we no longer calculate with 6 points, but only with 2 absolute values for the temperature difference instead of a gradient as withTTim, 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. This can be projected quite well, because I test the temperature difference on the two surfaces between which the paste is located.
Up to cA: 100 µm and below, the KPx performs a little better, only after that are the values of the Apex better. Interesting.
CPU emulation
Now I compare T1 of the reference with T1 of the gaming paste A. While the heater remains constant here, we already have sufficient thermal resistance in the copper reference block to simulate the CPU temperature and its differences with different pastes compared to the reference and depending on 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. But the TIMA5 test does. And we immediately see the first difference, where the KPx suddenly performs slightly better than the Apex at layer thicknesses below 150 µm. We are only talking about a difference of up to 1 K, but it is certainly reproducible.
34 Antworten
Kommentar
Lade neue Kommentare
Urgestein
Urgestein
Veteran
1
Urgestein
Mitglied
Urgestein
Urgestein
Veteran
Urgestein
Urgestein
Urgestein
Mitglied
Urgestein
Mitglied
Urgestein
Urgestein
Urgestein
Urgestein
Alle Kommentare lesen unter igor´sLAB Community →