If you’ve been following the occasional series of which this is part 5, overall, and part 4 about copper, you know that there are many different materials that have been used as the actual conductors for making cables and that not all of them are metallic. (A. J. Van den Hul, for example, uses carbon fibers in some of his products.) You also know that, while metals remain the overwhelming first choice of cable designers everywhere, there is still furious debate in some circles over which metal to use; with copper and silver being by far the most common for high-end audio applications.
Silver is unquestionably the more highly conductive (or, take your pick, least resistive) choice but, as I’ve attempted to show, resistance, while certainly one of the three factors (R = resistance, C= capacitance and L = inductance) that the conventional wisdom will admit to actually having some effect on a cable’s performance, is not necessarily the most important. In fact, as I said before, total resistance is easy to reduce (just use a thicker wire or a shorter one) and, at a maximum difference of only about 6% to 8% on the IACS (International Annealed Copper Standard) scale between silver and the cheapest copper, the difference in resistance isn’t all that much to begin with. This means, of course, that if there is, as many people believe, an audible difference between silver and copper cables, that difference must depend, at least in part, on something other than just resistance.
Another area where silver and copper are different is in the (still doubted in some circles) “tiny diodes” effect. This contends that, while thick films of copper oxide (CuO) are quite effective insulators, very thin films of it, such as are found on the surface of virtually any piece of copper that has ever, even briefly, been exposed to the atmosphere, or inside it, at the junctures of its internal crystals, are not insulators at all, but semiconductors — tiny diodes that resist current flow in only one direction.
Where this becomes important is that electrons passing through a conductor pass through a diode every time they pass through a crystal juncture. Also, because they “prefer” to travel in straight lines, they will jump from strand to strand in a stranded wire instead of just staying in the same strand that they started in and following it through the cable. What all this strand-to-strand jumping means is that every time there’s a jump, the electron has to pass through two tiny diodes (one on the surface of the strand it’s leaving and another on the surface of the one it’s jumping to). Do you remember what two diodes forms? Yup, a rectifier, and each time a jump takes place, a tiny amount of signal energy is, most people believe, rectified out of existence, ultimately and audibly affecting the total passage of signal through the cable.
A difference with silver is that silver oxide (Ag2O), while not as good a conductor as pure silver, is still a good conductor, and the oxygen-based “tiny diodes” effect may be eliminated. On the other hand, silver does not readily oxidize in air, and the bulk of the black tarnish that forms on it is not Ag2O, but silver sulphide (Ag2S), which, in thin films is, like copper oxide, semi-conductive. Because of that, an interesting problem arises: There may still be a “tiny diodes” effect with silver, but it may be sulphur-, rather than oxygen-based. And it may also, as with copper oxide, have the same semiconductor effect if thin films of silver sulphide form at the crystal junctures of the silver wires in a cable. In short, the difference may be no effective difference at all.
Another difference between copper and silver is claimed by certain high-end cable manufacturers (names purposely withheld) to lie in silver’s basic crystal structure. They describe this as “cubic lattice” and claim that, because of it, electrons have an easier time passing through silver than through copper or other lesser materials, and that this explains its greater conductivity. Rhodium, a metal increasingly used as the plating on connectors for high-end audio and AC power cables, is also claimed by many of those same people to have a cubic lattice structure and, according to them, that’s why rhodium-plated connectors sound better. I happen to agree that some of the rhodium-plated connectors I’ve tried have been quite good-sounding, but I can’t say that their crystal structure has anything to do with it. In all of the research that I’ve been able to do, I’ve only found three basic crystal structures for metals — Body-Centered Cubic (BCC), Face-Centered Cubic (FCC) and Hexagonal Close Packed (HCP) — and copper, silver and even rhodium all have FCC structures. Interestingly, other metals with an FCC crystal structure are aluminum, gold, lead, nickel and platinum, and they vary WILDLY in conductivity, from silver being best, at 106-108% IACS, to lead at 8.4% IACS. Rhodium, incidentally, has an IACS conductivity of just 38.4%, as compared to gold, the more popular connector-plating material, which is nearly twice as conductive, at 73.4% IACS.
That doesn’t really matter, though. In fact, as I have been saying throughout this whole metals-related portion of this series of articles, the whole issue of conductivity (and its opposite, resistance) is not, in my opinion, one of the most important things about cables or the things they are made of.
In fact — and here finally comes that thing that I have been promising to tell you about for so long, and that even I find a little difficult to deal with: Instead of the metal and its conductivity being the major determinants of electrical signal transmission, there’s a reasonable and substantial body of thought that suggests that they may have little or nothing to do with it!
Consider this: If you were to take two pieces of the same bare (uninsulated) wire, of exactly the same length, and cut off of the same spool, and measure the speed, as a percentage of the speed of light (300 million meters per second), at which an electrical signal passes through them (their velocity of propagation), you would find them to be exactly the same, to a good many decimal places. If, however, you were then to insulate those same two pieces of wire, each with the same length and thickness of a different insulating material (PVC and Teflon, for example), and again measure their velocity of propagation, you would find them to be significantly different.
How can that be? Isn’t insulation non-conductive? If so, then there can be no signal passing through either of the two insulating “jackets,” right? And if there’s no signal flow through the insulation, then all of the signal flow must be through the metal, right? So how can there be any difference in the velocities of propagation if it’s still the same two pieces of wire and the only things that have changed are the insulation?
There’s a considerable amount of well-reasoned thought by Maxwell, Faraday, Gauss and a number of lesser, but still formidable others, like Coulomb, that develops the idea that it is an “electrical field” surrounding the conductor and insulation that is the actual vehicle for the transmission of an electrical signal. As described by them, particularly in Maxwell’s equations, it is this field and its interaction with the objects within and around it that is the actual carrier of the signal and the determinant of the velocity of propagation — NOT the wire or the flow of electrons. If this is truly the case, though, then not only the wires, but actually the electrons that we are accustomed to think in terms of, assume an entirely new relationship to the signal and a great many of the conventional arguments and experience of the cable industry about how and why cables work may require re-examination and reinterpretation. If you are interested in this, an excellent article on Maxwell’s Equations can be found here. I will now refer you to it, and tell you only that next time in this occasional series on weird things about cables I’ll have something to say about connectors.
See you then!