7
\$\begingroup\$

I have a rather simple question about my fundamental lack of knowledge how is stereo signal transferred along 3 wire cable (L,R channels and ground)?

The problem with me understanding it is because as I understand the signal is AC voltage for both channels, right, but how are they then sharing the same ground? Wouldn't that produce crosstalk for L and R channels because they both use ground to carry the negative part of their AC signal?

What am I fundamentally misunderstanding?

\$\endgroup\$
7
  • \$\begingroup\$ you can avoid crosstalk with coating, galvanik decoupling and so on. I am not too sure how the circuit is build, but i imagine that this methods are used to prevent the signals from interfering themselves. \$\endgroup\$
    – Eggi
    Aug 22, 2016 at 8:29
  • 3
    \$\begingroup\$ @StjepanV: Please do not try to hijack someone else's question. \$\endgroup\$
    – Transistor
    Aug 22, 2016 at 8:49
  • 2
    \$\begingroup\$ @Transistor Ha? \$\endgroup\$
    – StjepanV
    Aug 22, 2016 at 8:56
  • 2
    \$\begingroup\$ @StjepanV: Sorry, the question appeared unrelated to the original post. I didn't notice that it was the OP. \$\endgroup\$
    – Transistor
    Aug 22, 2016 at 9:05
  • 1
    \$\begingroup\$ @StjepanV AC is not "two signal paths alternating with polarity." AC is transmitted across two wires where the relationship between the voltage of the wires alternates between positive and negative polarity. Each wire, on its own, doesn't really have a polarity. It's the relationships between them that matter. \$\endgroup\$
    – Cort Ammon
    Aug 22, 2016 at 15:59

3 Answers 3

12
\$\begingroup\$

The left channel is a loop on it's own connecting the left AC voltage to the Left (headphone) speaker.

Same for the Right channel, it is also an independent loop.

Now "by coincidence" the ground of both loops are connected. That does not change anything for the loops themselves as the current/signal flowing inside the loop cannot "escape", it stays in the loop.

The Left loop has its own signal and so does the Right loop. As long as there is only ONE connection between the loops, this works.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
17
\$\begingroup\$

There is slight crosstalk introduced. However, this is usually little enough to be negligible in actual consumer use.

If the ground wire were an 'ideal ground', that is absolutely zero resistance and inductance, there would be no crosstalk. That's because although the ground wire carries the current for both channels, because it's a perfect ground it always has zero volts at each end, so each of the left and right channel can reference an ideal 0 volt ground.

In practice, the ground wire has some resistance, let's say it has 0.1 ohms. If that cable is feeding a pair of 32 ohm headphones, that ratio is 320:1, or roughly 50dB, approximately the level of the crosstalk that would be generated. That is a better channel separation than vinyl records or multiplexed FM ever achieve, and once you put two loudspeakers in the same room, is anybody really worried about better than even 10dB?

If that cable were feeding another amplifier, rather than headphones, the input impedance would be much higher than 32ohms, and the ratio better, that is, practically perfect.

In studio work, all signals are carried individually.

\$\endgroup\$
9
  • \$\begingroup\$ I think that's what my second paragraph says, I even explain why, but I haven't bolded it up. \$\endgroup\$
    – Neil_UK
    Aug 22, 2016 at 9:03
  • \$\begingroup\$ It does, my bad. \$\endgroup\$ Aug 22, 2016 at 9:07
  • \$\begingroup\$ In studio work, all signals are carried individually. -- even better... balanced. \$\endgroup\$
    – Majenko
    Aug 22, 2016 at 9:38
  • 1
    \$\begingroup\$ As a side note, in headphones/earbuds some crosstalk is desirable (at least IMO), because most mixes are made for speakers. The air between the speaker and the listener's head introduces lots of crosstalk between the left and right channel. In headphones, this does not happen, leading to exaggerated stereo separation. \$\endgroup\$
    – Dampmaskin
    Aug 22, 2016 at 10:23
  • \$\begingroup\$ The numbers here are on the pessimistic end. A quick check with my headphones shows a cable resistance of no more than 0.01 ohms (ohm-meter shows 0, and claims to be able to show a resistance of 0.01 as non-zero) and a nominal impedance of 600 ohms. That works out to cross-talk being around 95-96 dB down (or better). \$\endgroup\$ Aug 22, 2016 at 13:27
1
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

Figure 1. Stereo headphone wiring. The common return from the speakers is connected to the jack sleeve which will connect to GND when plugged in.

There isn't really a problem. As long as the return conductor resistance is low relative to the resistance / impedance of the speakers then the speakers lower terminals will be held at GND.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.