Using Measurements to Clear Up the Cable Controversy

Using Measurements to Clear Up the Cable Controversy

Brent Butterworth

When I wrote my original article investigating whether the effects of speaker cables on speaker performance could be measured, I showed that changing speaker cables could have audible effects on the sound of a system.

For that test, I used mostly extreme examples: for example, a 24-gauge cable versus a 12-gauge cable. A lot of readers wondered what kind of difference I’d measure if I compared a generic 12-gauge cable to a high-end speaker cable. I wondered, too.

So I took what high-end cables I had, borrowed some really high-end cables from a couple of friends, and repeated the test.

Just to recap the testing method: I used my Clio 10 FW audio analyzer and MIC-01 measurement microphone to measure the response of one of my Revel Performa3 F206 speakers in-room. The in-room measurement was required to assure that there’d be no significant environmental noise. Yes, the in-room measurement shows a lot of effects of the room acoustics, but that didn’t matter because here, I was looking only for the difference in the measured result when I changed cables.

And just to recap the theory behind this: A speaker’s drivers and crossover components act as a complex electrical filter that’s tuned to give the speaker the desired sound. Adding resistance, in the form of a more resistive speaker cable, will change the frequencies at which the filter works and thus change the frequency response of the speaker. If the cable adds significantly more inductance or capacitance to the filter, then that, too, can affect the sound.

Test 1: AudioQuest vs. QED vs. 12-Gauge

Brent Butterworth

In my tests, I measured the effects of different high-end cables in 10- to 12-foot lengths and compared them to the measurement with generic 12-gauge speaker cable. Because the measurements were in most cases so similar, I'll present them here three at a time, with two high-end cables vs. the generic cable.

The chart here shows the generic cable (blue trace), AudioQuest Type 4 cable (red trace) and QED Silver Anniversary cable (green trace). As you can see, for the most part the differences are extremely tiny. In fact, most variances are within the normal, minor measurement-to-measurement differences you get when doing measurements of audio transducers due to trace amounts of noise, thermal fluctuations in the drivers, etc.

There is a small difference below 35 Hz; the higher-end cables actually produce less bass output from the speaker below 35 Hz, although the difference is on the order of -0.2 dB. It's highly unlikely this would be audible, due to the ear's relative insensitivity in this range; to the fact that most music doesn't have much content in this range (for comparison, the lowest note on standard bass guitars and upright basses is 41 Hz); and because only large tower speakers have much output below 30 Hz. (Yes, you could add a subwoofer to go that low, but almost all of those are self-powered and thus wouldn't be affected by the speaker cable.) You'd hear a larger difference in bass response by moving your head 1 foot in any direction.

I didn't get a chance to measure the electrical properties of the AudioQuest cable (the guy needed it back suddenly), but I did measure the resistance and capacitance of the QED and generic cables. (The inductance of the cables was too low for my Clio 10 FW to measure.)

Generic 12-gauge
Resistance: 0.0057 Ω per ft.
Capacitance: 0.023 nF per foot

QED Silver Anniversary
Resistance: 0.0085 Ω per ft.
Capacitance: 0.014 nF per foot

Test 2: Shunyata vs. High-End Prototype vs. 12-Gauge

Brent Butterworth

This next round brought out much higher-end cable: a 1.25-inch-thick Shunyata Research Etron Anaconda and a 0.88-inch-thick prototype cable that's being developed for a high-end audio company. Both appear thicker because they use woven tubing to cover the internal wires, but still, they're both heavy and expensive. The Shunyata Reserach cable goes for about $5,000/pair.

The chart here shows the generic cable (blue trace), the Shunyata Research cable (red trace) and the unnamed prototype high-end cable (green trace). Here's the electrical measurements:

Shunyata Research Etron Anaconda
Resistance: 0.0020 Ω per ft.
Capacitance: 0.020 nF per foot

High-End Prototype
Resistance: 0.0031 Ω per ft.
Capacitance: 0.038 nF per foot

Here we start to see some differences, especially above about 2 kHz. Let's zoom in for a closer look ...

Test 2: Zoom View

Brent Butterworth

By expanding the magnitude (dB) scale and limiting the bandwidth, we can see that these bigger, fatter cables do produce a measurable difference in the speaker's response. The F206 is an 8-ohm speaker; the magnitude of this difference would increase with a 4-ohm speaker.

It's not much of a difference -- typically a boost of +0.20 dB with the Shunyata, +0.19 dB with the prototype -- but it covers a range of more than three octaves. With a 4-ohm speaker, the figures should be double, so +0.40 dB for the Shunyata, +0.38 dB for the prototype..

According to the research cited in my original article, low-Q (high bandwidth) resonances of 0.3 dB magnitude can be audible. So by switching from a generic cable or a smaller-gauge high-end cable to one of these larger cables, it is absolutely, definitely possible that a difference could be heard.

What does that difference mean? I don't know. You might or might not even notice it, and it'd be subtle to say the least. I can't speculate on whether it would improve or degrade the sound of the speaker; it would elevate the treble, and with some speakers that would be good and others it would be bad. Note that typical absorptive room acoustics treatments would have a bigger measured effect.

Test 3: Phase

Brent Butterworth

Out of sheer curiosity, I also did a comparison of the degree of phase shift caused by the cables, with the generic cable in blue, the Audioquest in red, the prototype in green, the QED in orange and the Shunyata in purple. As you can see above, there's no observable phase shift except at very low frequencies. We start to the see the effects below 40 Hz, and they get more visible down around 20 Hz.

As I noted before, these effects probably wouldn't be very audible for most people, because most music doesn't have much content at such low frequencies, and most speakers don't have a lot of output between 30 Hz. Still, I can't say with certainty that these effects would be inaudible.

So DO Speaker Cables Make a Difference?

Brent Butterworth

What these tests show is that the people who insist you can't possibly hear a difference between two different speaker cables of reasonable gauge are wrong. It is possible to hear a difference by switching cables.

Now, what would that difference mean to you? It'd definitely be subtle. As the blind comparison of generic speaker cables we did at The Wirecutter showed, even in the cases where listeners can hear a difference between cables, the desirability of that difference may change depending on the speaker you use.

From these admittedly limited tests, it looks to me like the big differences in speaker cable performance are due primarily to the amount of resistance in a cable. The biggest differences I measured were with the two cables that had substantially lower resistance than the others.

So yes, speaker cables can change the sound of a system. Not by a lot. But they can definitely change the sound.

More From Us