Giz Explains: Why Analog Audio Cables Really Aren't All The SameWhile I've clearly stated that, despite Best Buy's 57 overpriced "choices," HDMI cables pretty much all do the same thing, this isn't the case with analog audio cables. Here's the physical explanation why:
While there aren't a lot of parts to audio cables, there are more than you might think, especially on higher-end cables. Their goal, like most wiring, is to transmit an electrical signal from point A to point B without losing signal strength or clarity. But because this is the real world and not fantasy land, there will always be some kind of loss, and when the signal is analograther than digitalthere's less of an all-or-nothing tradeoff. With the HDMI cables we tested, most could pass the required 1080p video and audio signal with no noticeable blips, but with analog, there are infinite gradations of signal that could actually make it through.
What gets in the way of the signal? A cable has several key electrical properties that differ from manufacturer to manufacturer, and can be manipulated by the way a cable is constructed. Because that manipulation usually involves performance tradeoffs, cable makers themselves have to ask two questions:
1. How much can I minimize the signal loss overall?
2. If I have to compromise, what do I sacrifice and what to I preserve?
I asked my friend David Kay, an editor at Audio Junkies, to help me sort out this mess, and we recently walked through all the electrical properties that have an impact on an audio signal as it passes along a cable. Here's the gist of our discussion:
Resistance: How much opposition the cable's own length, shape and material impose on the signal passing through. Since the opposite of resistivity is conductivity, a better conductor material used to make a wire inherently means lower resistance. "Exceptionally pure and exceptionally drawn copper or silver is much more conductive than the cheap stuff," says Kay. "Some companies use rare earth materials, like palladium, which is supposedly more conductive than silver." The word impedanceoften used in audio terminologyis essentially a derivative of electrical resistance.
Measured in ohms, or rather milliohms, resistance is usually referred to by the letter R. The fewer milliohms of resistance, the better time you're gonna have.
When wire makers are casting or drawing their metals into long strands, they take care to keep oxygen out of the process, and the more expensive cables take care to keep the metal hot while working with it, to avoid the build up of crystals. It is believed that oxygen and crystal buildup in the metals affect sound quality, because they add to the resistance. One of the most celebrated casting methods for super-pure, high-conductivity copper is called the Ohno Continuous Casting process, or OCC.
Capacitance: "The ability of a body to hold an electrical charge," as Wikipedia puts it so concisely. A capacitor is basically two conductors separated by an insulator. If that sounds suspiciously like every speaker wire you've ever seen, then you're on to something. Modern capacitors found in consumer electronics are purpose built to hold charges, like smart, fast mini batteries, but even primitive cables can hold some kind of electrical impulse. If that happens, it's not going to get where it's going on time. And time, as it happens, is important when listening to music.
Capacitance, indicated by a C, is measured in Farads. In the audio world, usually that's picoFarads (pF), the fewer the better.
Inductance: Sparing you the nasty official descriptions, in this case, it's the potential for a conductor (wire) to generate an electromagnetic field based on the electricity flowing through it. Too much of an electromagnetic field can technically screw with the audio signal. (Britannica.com actually has some decent chatter on inductance, if you dare.)
The Romulus to capacitance's Remus, these two are always at odds. Even if you can bring 'em both down, cable makers always reach a point where they have to make a tradeoff between capacitance and inductance. Says Kay, "The way you arrange wires in your cable, by twisting them, and by deciding how far to space the conductors apartthat all has an enormous effect. You want a balanceyou don't want ultrahigh readings of one to get low readings of the other."
Inductance is denoted with an Lapparently as a shoutout to some long-dead physicist. You track induction in Henry units (not making this up), also a shoutout to some dead dude, and yeah, they measure inductance in audio cables with microHenrys (µH or uH).
The Dielectric, aka Insulation: One of the biggest disputes in high-end cable is how to insulate your conductors. Lay people tend to think that the plastic coating around a wire is there to keep the electricity from getting out, or other electricity from getting in, and while that's mostly true, the physics argument says that the signal itself travels between the conductor and the insulator, almost always referred to in cable science as the dielectric. The wave is technically jumping along the wire, bouncing off of each insulated wall as it goes. Because of this, some warrant that an insulator shouldn't be too absorptive. Cable makers tend to frown upon PVC, and smile at Teflon. In the middle, says Kay, are things like polyethylene foam. Teflon is expensive and harder to extrude over copper or silver wire, hence its price.
There's more to insulation than a mere coating. In fact, some cable makers coat each individual strand of copper in what's called a Litz coating, in order to ensure the longevity of the cables, according to one of the makers. The strands themselves aren't microscopicthey're usually thick by typical standardsbut inside each "wire" you find a bundle of smaller individually wrapped wires. You can imagine this is a bit of a pain to work with if you're stripping wire yourself, but some people do like it.
Skin depth: As Stereophile's John Atkinson put it back in 1995: "There is an optimum conductor diameter for audio-signal transmission." This has to do with research conducted by a University of Oxford professor named Malcolm Omar Hawksford (bonus points for cool multicultural name), who said that lower frequency signals take longer to travel through thicker wire, creating an effect known as "time smearing." The solution was to bring the diameter down to somewhere between 0.5mm and 1mm per strandthat is, 24 to 18 AWGin order to ensure that "a uniform current flow across the conductor is to be maintained over the audioband."
In order for cable makers to avoid problems resulting from skin depth, some have resorted to making hollowed-out or irregularly shaped cableslike the hollow oval cable at left from Analysis Plusor arranging skinny cables around a hollow center.
Prof. Hawksford also corroborated a few other points that had been made (and that I mentioned above):
"The dielectric supports the majority of the signal during its transportation along the cable"
"Stranded conductors without individual [Litz] strand insulation appear to be a poor construction"
Network Boxes: Some speaker cables from certain manufacturers have "network" boxes on them, electronic boxes that are used to ensure electromagnetic noise from the system itself doesn't get through, effectively filtering low frequencies that wouldn't be in the music. Higher-end cables let you select the impedance, that is, tune the cable to show preference for low, mid or high frequencies. (At left, MIT Shotgun speaker wire with network box.)
Shielding: You may note that I haven't yet bring up shielding, even though it's a popular topic in audio cables. It's because you generally only shield interconnects, which are constantly carrying low-energy signals between heavy electrical components. Those weak signals are easy prey to external electromagnetic interference, while the higher-voltage stuff running through speaker cables tends to be impervious to that outside noise.
Shielding comes in different formats, usually all at the same time, in order to cover different electromagnetic frequency ranges, from the low end, protected by a braided mesh of aluminum, copper or even silver-plated copper, to the high end, kept out by solid sheets of aluminum or mylar. An additional layer of carbon-filled (conductive) PVC is also used to block certain electromagnetic frequencies and to cut down on the tendency of some interconnects to be "microphonic," picking up vibrations that turn to sound. (At left, Cardas' interconnect with three layers of shielding, Litz-coated individual strands of cable, and even a hollow centerlots of principles at work here.)
Kay says that, because of this vibration issue, there are even more unusual shields coming to market, including wires surrounded by beads and sometimes even fluids.
WHOA WHOA WHOA, BUT CAN YOU HEAR ALL THAT CRAP!?!??!!!!
The question is, how does this actually lead to differences in sound quality? Can people actually hear differences in cables when they are optimized in accordance with the physical properties I talked about up above?
When proving his theory of skin depth, Professor Hawksford wrote, "I am not trying to say that this effect is necessarily significant, only that an error component is predicted by our theory and is shown by the measurements to exist." I think that sums up where the real division lies in the audio cable debate. Some people say they can hear it, others refuse to buy that these proven tweaks of electricity and electromagnetism are audible.
You're not going to hear it on a $350 home-theater-in-a-box system that came with speakers thrown in. But if you spend a few grand per speaker, and a fair amount on a receiver, you just might stand a chance of hearing some differences. If you spend $100,000 on a home-theater setup, well, you've probably got the cash to buy any audio cables you want, so I'm not sure I give a damn what you buy.
And as I mentioned up top, we're talking about analog cable here. It's true that anything passing electrical signalsbe they analog or digitalwill encounter these same issues, but as we've seen, the heavy demands of 1080p video mean you either see the picture or you see something revoltingthere's less of an opportunity for nuance, which makes cheaper digital cablesespecially ones with a guaranteea more sensible purchase.
This isn't necessarily a validation of extravagant spending when it comes to speaker cables and interconnects, but it does acknowledge that even skeptics can hear the difference between a $200 speaker cable and a $1000 speaker cable, given the right (generally very expensive) sound system.In fact, many say that higher-end systems reveal the qualitative difference between cheap and high-end cables better than they reveal any variation among various expensive ones. That is, the jump from $20 to $1000 is much more clearly audible than the jump from, say, $2500 to $5000.
I get this, because when tasting wine, my ignorant ass can correctly identify the differences between a $9 white Burgundy and a $25 one, but I may have a lot of trouble sussing out what makes a $100 white Burgundy any better than that same $25 bottle. If you put a $100 bottle and a $1000 bottle in front of me and ask me to tell the difference, I'd probably just walk away. (Actually, I'd grab them and run, but only to sell 'em on eBay.)
I'm not going to leave you with some direct answer. In fact, we've reached a nice pausing point where you can yell at each other (or me) about all the reasons you think all this stuff is real or total nonsense. The good news is, we're cooking up a version of the age-old analog cable test, to see what we can learn for ourselves, both with analytical gear and our very own ears. Meantime, please discuss this in comments, and if you have any suggestions for your own tests, I'd love to hear them below or via email.
Also, there's a damn good chance you know more about electrical engineering than I do, and that you'd like to elaborate on or maybe even correct something I discussed above. Just don't be a dick about itit's tricky to put this stuff into terms every interested party can understand, and in doing so, stuff gets left out. By all means share your knowledge, but know that this is just the beginningnot the final word.
