The influence of shielding and insulation on interlinks

Measurements and conclusion

We naturally wanted to see how the various insulation and shielding materials affected reproduction as well as whether we could measure anything. Measurements were done in line with the earlier mega-test of interlinks. So we took the pulse generator and Tektronix scope as well as the Sourcetronic LCR, the ‘black’ measurement box as well as the spectrum analyzer.

LCR Measurements

It is quite extraordinary how differences are still clearly measurable. The impedance differences are incredibly small. But then again, we expected that. Inductance, however, is a remarkable one: at the copper shielding, we see that the inductance drops towards 300 kHz. The rest of the field remains nicely straight. Inductance is decent on average, but not super low for a 0.5m cable. Still, the difference between the lowest and highest is 80 nH.

Looking at capacitance, the differences are incredibly small, but consistent. The RF cloth adds capacity. Aerogel without the tubes has the lowest capacitance. The difference between the lowest and highest is about 3pF. This is negligible.

Capacitance and inductance again express themselves in impedance. There too, small differences are measurable. We know that impedance has no influence on the musical reproduction of interlinks.

Scope measurements

Noise

Here we can clearly see differences between no insulation, insulation and shielding. What is noticeable is that both copper and RF cloth even out the noise spectrum. Aerogel doesn’t do much. In fact, it seems to get messier. RF cloth clearly wins in terms of noise level. It is also, as mentioned, a lot more even accross the spectrum.

Propagation time

The propagation time is the same for all cables. Now these are cables of 0.5m, so really we can’t measure it well: the differences are too small with cables of 0.5m. It is funny to see how the pulse and reflections look though…

Spectrum

Again, we see subtle differences. Shielding definitely affects the spectrum. And the shape of the pulse in a way. With longer cables we would see it more clearly, but even with 0.5m we see the spectrum change. It’s not so much that we see it getting “drier,” but we definitely see shifts in the peaks and with the copper foil we see that a peak disappears.

It’s hard to measure with such short cables, but we can certainly say that the spectrum is shifting subtly. And pointing to the listening tests, even there the differences were subtle, but audible. So that does rhyme with the subtle but reproducible shifts in the spectrum.

Response

Again, small differences that are reproducible. Realize that we only changed the insulation and shielding materials. Nothing else! RF cloth gives the most deviation towards higher frequencies. Which is not very crazy. By the way, we are talking about 0.1 dBm at 10 MHz. In most cases then, we are talking about a margin of error….

Conclusion

It is and remains fascinating what cables do. Conductors have influence, geometry has (considerable) influence and so do insulation and shielding materials. All in all, an engineer can give a cable a clear signature by playing with it. And we noticed this during the big test of interlinks!