For the less experienced: "Current" is movement of electrons. This movement is accompanied by an electric field and a magnetic field. We usually think of the magnetic or electric fields as forcing the electrons to move, but a stray electron blasting through the area (even in a vacuum) will be associated with an electric and a magnetic field. When we talk about "Induced hum" in our electronics, we usually are implying magnetic field coupling where an external magnetic field generates current in the wire that adds to the desired signal.

In an electric generator we are causing relative movement between a magnet and a wire. This movement forces electrons to move in the wire -- current. If there is no movement, there is no current. (and no voltage)

In a power transformer we are using the alternating magnetic field created by electrons moving (current) in the primary winding, to induce a magnetic field into the secondary winding (forcing electrons to move in the secondary -- current), and if we are clever about things, the secondary voltage will be more convenient in some context.

If your signal wire passes through a region with a relatively strong magnetic field (possibly created by a motor, power transformer, or current carrying power wiring) this magnetic field will create a low grade transformer that will add hum to your delicate little signal.


Again, for the inexperienced: In order to understand "Ground Loop" (by the way, a "loop" is simply a piece of wire or a circuit trace that goes from here to there) let's define a perfect four terminal device, an amplifier, where the output is simply a multiple of the input. Let's keep things simple and label the input and output terminals as "i+", "i-", "o+", and "o-". Our theoretical amplifier is buried deep inside one of the boxes that we mount in our rack. Actually, there might be multiple "amplifiers" mounted inside our box. (Phono preamp, power amplifier, RF amplifier, tone control, etc.)

We usually call the zero point in our circuit "ground". Unfortunately, we are very sloppy with this term and "ground" could be at the "i-" terminal of our theoretical amplifier, a point on a circuit board, the shield on a coax cable, a connection to the chassis, the third wire in a power cord, a rod driven into the soil, or any other reference point we might use to measure voltage.

As you can imagine, the sloppy use of the term "ground" will lead to trouble.

Remember, whenever current is flowing in a conductor, there will be a voltage drop due to resistance. When we are modeling circuits we often draw a little generator symbol in this conductor to remind us that there is a voltage difference from end to end when we use the same reference point while measuring the voltage on both ends of the wire and this "generator" is the result of some physical process. (Actually, there is a voltage drop in the leads of the voltmeter too)

By the way, it can be very helpful if you draw a little generator symbol in every little scrap of wire, shield, circuit trace, transistor lead, etc. in your project. A physical process will be driving each one of these little "generators". Of course, many of these voltages will be insignificant, but many will be quite significant.

Now, let's imagine a printed circuit board where our delicate little signal enters on one side and the shield is connected to the board's "ground" trace at this point, while our ideal amplifier is located on the other side of this board. Next imagine a sloppy bit of design work where a fan is also connected to this "ground" trace and the fan is located on the same side of the board as our ideal amplifier and amplifier and fan share the same power supply. (We now have a little noise "generator", associated with the fan, introducing a "signal" into our "ground") Meanwhile, our signal travels straight to the ideal amplifier's "i+" terminal. We have created a classic grounding issue because the current for the fan is added to the signal presented to the "i-" terminal of our ideal amplifier and our amplifier output is a multiple of the voltage drop in the "ground" caused by the fan current plus a multiple of our signal. If the design is really sloppy, the ground signal might swamp the audio signal.

There have been experiments where the same circuit is built by different teams and the resulting equipment performed differently due to one team's better layout and management of "grounds".

---

The above is a quick overview.

The is enough meat in the "ground" drama to keep specialists employed for a career -- just straightening out subtle and gross oversights by others.

---

A consideration for our shielded wire: The shield is used to protect the coaxial central wire from stray electric and magnetic shields. Actually, the shield is not very effective at protecting against magnetic fields. (this is why we don't want to run our signal cable beside high current power wiring) The idea is that the shield represents an easier path for the fields than the insulator around the central wire and the fields will tend to remain in the shield and stay away from our signal.

Now consider the idea that the shield is connected to the chassis at both ends of the cable and one can measure a voltage difference between the chassis. Now, we have current flowing in the shield caused by the different chassis potentials. You might come across this sort of situation labeled as "circulating current" because it is present with or without our audio signal. This is really bad because there will be current flowing in the shield due to this voltage difference. Since there is a magnetic field associated with this shield current, a signal is induced into the central wire that the shield should be protecting. (never mind that there could also be messy "ground" management in either box that contributes to our troubles) Because this chassis to chassis voltage difference is usually associated with the power line, we experience "hum".

This suggests a quickie technique that many of us have used: Break the shield connection at one end of the cable and make a chassis to chassis connection outside of the shield. This will eliminate chassis to chassis current in the shield. I'm not suggesting that this is universally the best technique, but sometimes it will result in a quickie miracle.

BTW, this is a technique used by some premium audio cables.

---

Whew! I'm out of time at the moment, maybe someone will follow on and explain why "balanced line" is so helpful when dealing with and preventing noise issues. And, why that missing cable system "ground" can be so bothersome.