Speaker cable: what resistance actually costs

Cable is the easiest subject in audio to argue about, because the numbers involved are small in absolute terms and the arguments are usually about something else. This is the arithmetic: a length, a cross-section, and the resistance they produce.

The resistance is then a divider against the loudspeaker, and the divider sets two numbers that are worth knowing — the damping factor the amplifier sees at the terminals, and the fraction of a decibel the cable costs. Both are computable, and both are usually smaller than the argument implies.

Two conductors, one loop

Resistance is proportional to length and inverse to cross-section, and the length that matters is the round trip. A 4 m run of 2.5 mm² copper is 8 m of conductor, and at 1.68×10⁻⁸ ohm-metres that is 54 milliohms.

For scale: a voice coil's DC resistance is typically 5 to 7 ohms, and the amplifier's own output impedance might be 0.02 to 0.1 ohm. The cable sits between two much larger terms on one side and a much smaller one on the other, which already tells you where it can and cannot matter.

Damping factor is a ratio, not a specification

Damping factor is load impedance divided by the total resistance in series with it. An amplifier with 0.05 ohm output impedance driving 8 ohms gives 160. Add the 54 milliohm cable and it becomes 77.

That looks like a large change and is not, because the number that actually damps the cone is the total resistance in the loop — including the voice coil's own 6 ohms. Against that, the difference between 5 and 10 percent of the total is a rounding error in the electrical Q of the driver, which is why the audible effect of damping factor above about 20 is essentially nil.

The loss is small, and it is a divider

The cable and the load form a voltage divider, so the loss depends on their ratio rather than on the cable alone. Into 8 ohms, 54 milliohms costs 0.058 dB — about a seventieth of the smallest change most people claim to hear.

Two things make it worse without changing the cable. The load is not 8 ohms at every frequency: at the impedance minimum it might be 5, which raises the loss by half. And the loss is a power loss, so it is the same current through more resistance — a long thin run into a 4 ohm load is the case where the arithmetic finally produces something visible.

Where cable actually matters

Three cases, and none of them are the usual argument. Long runs in a large room, where the length term dominates: 20 m of 1.5 mm² is 0.45 ohm, which is 5 percent of an 8 ohm load and worth fixing. Low-impedance loads, where the same absolute resistance is a larger fraction. And any case where the resistance is in series with a passive crossover, because there it shifts the filter's own alignment rather than just costing level.

There is a fourth case worth mentioning because it is measurable: series resistance raises the driver's electrical Q, and therefore its Qts. Feeding a closed-box design 0.45 ohm of cable and crossover DCR can move Qtc enough to change the box. That is a real mechanism, and it is the one to check before blaming the cable for a bass problem.

Gauge, in the numbers people actually use

• 0.75 mm² — about AWG 18. Fine for a metre into 8 ohms.
• 1.5 mm² — about AWG 15. The usual domestic minimum.
• 2.5 mm² — about AWG 13. 54 milliohms over a 4 m run.
• 4 mm² — about AWG 11. 34 milliohms over the same run, which is past the point of mattering.

Five mistakes worth avoiding

1. Using the one-way length. The loop is twice the run, and halving the resistance by accident is the most common error here.
2. Comparing damping factors measured differently. An amplifier's specification is at its own terminals; the system figure includes the cable.
3. Ignoring the impedance minimum. The loss is worst where the load is lowest, not at the nominal figure on the box.
4. Blaming cable for a bass problem. Check Qts and the box first; series resistance is a real effect but a small one.
5. Buying gauge instead of checking contacts. A corroded terminal is tens of milliohms that no amount of copper downstream will fix.

Summary

Double the length, divide by the area, and you have the loop resistance. Compare it with the load to get the loss, and with the amplifier's output impedance to get the damping factor. Both answers are usually small enough that the honest conclusion is "this cable is fine", which is a more useful result than another argument.

Values shown are engineering aids rather than measurements; see the disclaimer and the tool index.