Expensive and cheap speaker cables: what's the difference? Part 1

Note translator: the text uses the classification of cables in accordance with the American system of marking the thickness of the wires. Specify cable specifications in accordance with the international standard for wire gauges IEC 60228 (AWG) here .

The recent speaker cable test I conducted for Geoff Morrison on The Wirecutter made me re-think about a very controversial issue, which I thought was forever left my head.
')
Later, I told Alan Devantier (Allan Devantier), manager of acoustic research at Harman International (who produce Harman Kardon receivers, JBL and Infinity speakers, as well as products from many other audio brands) that we had in mind. We have dealt with all the details about whether it is possible from a technical point of view to demonstrate that in quite extreme conditions, speaker cables can make noticeable changes in the sound of your system.

But first, I’ll make a reservation that I have no clear beliefs about speaker cables. I conducted “blind tests” for the Home Theater magazine, in which participants showed clear preferences for certain cables, but I myself didn’t do it myself.

Two different opinions confused me. The first point of view was the belligerent confidence of some audio publications (in particular, Stereo Review , which has now become Sound & Vision, and The Audio Critic barely alive) that the speaker cables do not affect the sound. The second was the long-term, complex and stormy descriptions of differences in the "sound" of cables, expressed by professional audio reviewers. It seems to me that both sides are upholding established positions, instead of honestly and unbiasedly getting to the bottom of the truth.

If you are interested in knowing what I use, I’ll say that I use Canare speaker cables - these are typical 14-gauge four-core cables for laying long tracks, and a few other cables. I have to add that in more than 20 years of speaker reviews and testing of different models priced from $ 50 to $ 20,000 per pair, I received a question only once from a manufacturer who was concerned about which speaker cables I used.

Alan's research

Devantier became interested when I began to talk about how a speaker cable could, in theory, change the frequency response of a speaker. In fact, each speaker is an electrical filter, that is, a combination of resistance, capacitance, and inductance, tuned (hopefully) to produce the best quality sound. If you increase the resistance, capacitance or inductance, then you change the filter parameters, and as a result - the sound of the speaker.

An ordinary speaker cable does not have any significant values ​​of capacitance and inductance, but its resistance changes, especially with thin cables, because, all other things being equal, the thinner the cable, the greater the resistance.

I will allow Devantier to continue. He begins by referring to the research of Floyd Toole, his former colleague, and Sean Olive, already a current colleague at Harman, who once worked on the National Research Council of Canada:

In 1986, Floyd Thule and Sean Olive published a study on the audibility of resonant vibrations. They found that listeners are most sensitive to low-Q (high-throughput) vibrations. Under the right conditions, one could hear mid-frequency peaks of just 0.3 decibels (dB). Since the impedance of the speaker changes with frequency, the resistance of the cable core to direct current becomes a very important parameter. The following diagram shows the maximum allowed cable length, chosen so that the signal amplitude fluctuations caused by the cable resistance are maintained below 0.3 dB. This scheme assumes that the minimum impedance of the speakers is 4 ohms, and the maximum is 40 ohms, and the cable resistance is the only influencing factor. The circuit does not include capacitance and inductance, which can only make things less predictable.

"From this table it should be clear that in some cases the cable and the speaker can interact and cause audible resonant vibrations."

Brent measurements

“You know that you can measure it,” Alan told me, making a finger movement, hinting that it was a statement, not a guess.

I measured the frequency characteristics of the speakers since 1997, but I always used just a good, big, thick cable to connect the speaker being tested to an amplifier, a cable I was sure of and did not affect the measurement accuracy. But what if I replace it with a small, cheap and "lousy" speaker cable? Can I measure the difference? And will it be a difference that can be heard?

To find out, I measured the frequency response of the Revel F208 speaker using the Clio 10 FW audio analyzer, using three different 20-foot (approx. 6 m) cables:

1. 12 gauge Linn cable, which I used to measure speaker parameters for the last five years or so;
2. Cheap Monoprice 12 gauge cable;
3. Cheap RCA 24 gauge cable.

To reduce ambient noise, I took measurements indoors. Neither the microphone, nor the speaker, nor anything else in the room did not move. I used a super-long FireWire cable, so I and the computer were in another room. In addition, I repeated each measurement several times to make sure that the ambient noise did not affect (to a large extent) the measurements. Why so carefully? I knew that I was measuring minor changes, if they could be measured at all.

Then I measured the characteristic with a Linn cable and divided it into the characteristics of Monoprice and RCA cables. This made it possible to construct a graph reflecting the differences in frequency characteristics caused by cables. After that, I applied 1/3 octave anti-aliasing to make sure no ambient noise remained on the recording.

Devantier was right - I can measure the differences. And as you can see on the graph, the results of the two 12 gauge cables differ slightly. The biggest change was an increase in the signal level of 0.4 dB in the frequency range from 4.3 kHz to 6.8 kHz.

Is it possible to catch these changes? Maybe. Deliver it any inconvenience? Probably not. To objectively evaluate this, I will say that I usually notice a change of 20-30% when I examine a speaker with and without a protective grille.

But using a 24 gauge cable had a striking effect. To normalize the measured response curve level, I had to increase it by 2.04 dB, so that I could compare the measured data with the curve obtained using a Linn cable. 24 gauge cable resistance also had an effect on frequency response. For example, the cable cut the bass at frequencies between 50 Hz and 230 Hz by a maximum of -1.5 dB (at 95 Hz), cut off the average frequencies of 2.2 - 4.7 kHz by at most -1.7 dB at frequency 3, 1 kHz, and reduced the upper frequencies 6 - 20 kHz in size to -1.4 dB at a frequency of 13.3 kHz.

Can this be heard? Definitely. Is it very noticeable? Highly. Do you like the sound with a thin cable or with a thick one? I dont know. Regardless, our last recommendation for using 12 or 14 gauge cables was pretty wise.

This is a pretty extreme example. Although there may be some exotic high-resistance acoustic cables, almost all cables with a thickness of about 14 have a rather low resistance, so all sound anomalies that they introduce should be inaudible. But I consider it important to note that I measured small recurring differences in the characteristics of the two cables, similar in size and structure. In addition, note that the Revel F208 has an average resistance of 5 ohms (personally measured). The resulting effects would be more pronounced with a speaker whose resistance is 4 ohms, and less pronounced if its resistance is 8 ohms - such dynamics are the most common.

So what kind of lesson can we learn from all this? Basically, we learned that you should not use thin cables in any system that requires high-quality sound. And also the fact that you probably should not rush to lash out at people who say they can hear the differences in the sound of the speakers with different types of cables. Of course, many of them are grossly exaggerating these effects, as are the companies that produce such cables. But the calculations and measurements we conducted here suggest that it is quite possible to hear something like that.

[ Part 2 ]

In one of our podcast releases, we already talked about digital and analog sound from the point of view of its basic aspects, along with Anatoly Dmitrievich Arsyonov, Ph.D. (physicist by training, an expert in IT and digital sound, an engineer at F-Lab).