Over the years that I've worked in electronics, I've noticed several common misconceptions about the use of the word "ground." Some of these misconceptions comes from our desires to shorten nomenclature designations to fit the tiny box on our wiring documentation. A large part of the problem comes from the valid need to simplify circuitry for analysis and study, particularly in school. But I know the differences in how and why a wire is connected to earth produces very different effects and requirements.
In a perfect world (please no snickers 
) the use of the word "ground" would be limited to the path of lightning strikes, other unintended currents and a return path for antenna based radio frequency emissions. Return would designate the path for intended currents. This designation differentiation does exist in AC power distribution with labels Neutral and Ground. Also, above my head I happen to have the pin designations for the arcane NIM (Nuclear Instrumentation Module) package. Discarding the shield connection of the six coaxial connectors as grounds and the numerous chromated chassis contacts there are three additional pins that at some point connect to the power cord's ground pin; the power return ground, the high quality ground, and the 117 volts A.C. neutral.
This brings me to the title of this discussion. Frequently we forget that different things that are connected to ground need not be at identical potentials. While they can be at the same potential at a given moment in time, rarely will this true. As we all should remember, this happens from several factors; the IR drop across the connecting wire, ground loops and interference. Ground is not zero volts, this is a myth. Even when one uses superconducting material for part of the circuitry. The factors of inductance, contact resistance transitioning to room temperature conductors, the resistance of the room temperature conductors and any ground loop effects created will create a voltage potential. Predominantly, these differences can easily be made so small that they can be ignored. But "close enough" is not the definition of identical.
Using this approach to separate earth connections into two groups one can measure micro-amperes of current just a few meters away from a megawatt RF amplifier that is not the RF amplifier's interference signal. By separating return current loop paths from ground loop paths in one's circuit analysis, more accurate and reliable measurements can be accomplished. This does mean that frequently one wire will carry both a return and ground current. But recognizing the difference between the two leads to better signal integrity.
The Myth of Ground
Re: The Myth of Ground
