The idea that “dirty” power causes audio system noise problems has a nearly irresistible intuitive appeal – and there are dozens of companies ready to cash in on this widespread but mistaken belief.
For example, here is a quote from a well-known manufacturer of power conditioning products: “Today’s residential systems contractors face unprecedented challenges where high resolution, trouble-free operation is required.
From inducing AC ground loops, video hum bars, static bursts, damage from AC line surges and variable audio and video performance, comprehensive control and conditioning of AC power is no longer an option.“
In fact, the power line doesn’t cause ground loops at all—and no amount of power “cleansing” or “purification” will prevent them!
Obviously, if every highway were smooth as glass, our cars wouldn’t need suspension systems. But it’s simply unrealistic to expect such highways – we pretty much have to accept them as they are.
The same is true for the AC power line. It’s a utility used by all sorts of appliances and equipment – and it’s certainly not pristine. Further, the power distribution systems in our buildings unavoidably create small voltages and currents that can potentially contaminate our signals.
Therefore, we need “suspension systems” to isolate our audio signal paths from the power line. Any pathway that allows coupling between the two is the fundamental problem causing noise in sound (and video and computer) systems.
Even though this is demonstrably true, and based on real science, it’s often difficult to persuade folks that “bad” AC power isn’t to blame. Audio systems routinely suffer from hum and buzz even when AC power is pristine.
In unbalanced interconnections, the noise is usually coupled in the audio cables. It isn’t that the cables are poorly shielded; rather, it’s due to the basic properties of wires.
A simplified equivalent circuit of an unbalanced audio cable (Figure 1) shows that the shield, like any wire, has both DC resistance and inductance and that this inductance is magnetically coupled to the inductance of the center conductor, creating a kind of transformer.
The impedance of an inductance increases in direct proportion to frequency. Therefore, when current flows in the shield at frequencies below about 10 kHz, most of the voltage drop occurs across the resistance and very little across the inductance. It is this voltage drop that adds noise to the audio signal and is responsible for 99 percent of noise problems in unbalanced interfaces.
But, at higher frequencies, most of the voltage drop occurs across the inductance and, through transformer action, induces an equal voltage in the center conductor, thus reducing the coupling as frequency increases.
We may conclude that noise coupling in unbalanced interfaces is a significant problem only at audio frequencies. Balanced interfaces are generally immune to this coupling mechanism, but can fall victim to others.