I think part of the problem is that we assume that the obvious truths are obvious to everyone. Many of the texts and reference materials have deleted the fundamental explanations for why we do what we do. Even if we learn later that the methods were based on what 20/20 hindsight tells us are not quite right reasoning, it's still good to know the initial motivation behind the rule.

I've taught other subjects and I've lost count of how many times I've said something referencing X, and been confronted with 20-30 deer in the headlights because I assumed they already knew X coming in.

I think we need to get a little more thorough in the backstory.

I appreciate the effort. I wish we could have the discussion on voltage gradients around metal objects, but we can't and I don't want to start the entire conversation again. But the following is from IEEE 142.

2.2.8 Connection to Earth Earth is inherently a rather poor conductor whose resistivity is around one billion times that of copper. A 10 ft (3 m) long by 5/8 in (16 mm) diameter ground rod driven into earth might very likely represent a 25 ohm connection to earth. This resistance may be imagined to be made up of the collective resistance of a series of equal thickness concentric cylindrical shells of earth. The inner shell will of course represent the largest incremental value of resistance, since the resistance is inversely proportion to the shell diameter. Thus the central small diameter shells of earth constitute the bulk of the earthing terminal resistance. Half of the 25 ohm resistance value would likely be contained within a 1 ft (0.15 m) diameter cylinder (see 4.1.1).

For the same reason, half of the voltage drop resulting from current injection into this grounding electrode would appear across the first 0.5 ft (0.15 m) of earth surface radially away from the ground rod. If a current of 1000 A were forced into this grounding electrode, the rod would be forced to rise above mean earth potential by 25 000 V (1000 • 25). Half of this voltage (12 500 V) would appear as a voltage drop between the rod and the earth spaced only 0.5 ft (0.15 m) away from the rod. While this current is flowing, a person standing on earth 0.5 ft (0.15 m) away from the ground rod and touching the connecting lead to the electrode would be spanning a potential difference of 12 500 V. A three-dimensional plot of earth surface potential versus distance from the ground rod would create the anthill-shape displayed in Fig 36. The central peak value would be the rod potential (referred to remote earth potential), namely, 25 000 V. Moving away from the rod in any horizontal direction would rapidly reduce the voltage value. The half-voltage contour would be a horizontal circle 1 ft (0.3 m) in diameter encircling the rod.

and I have a question about the possible shock hazzard of touching the grounded surface of a system during the instant of a ground fault. If you have energized the bonded metal parts (with what ever potential) and you touch this metal. What will you also have to touch or be in contact with to be a path for current to flow on? certainly if you touch another bonded metal object it's not likely to be of a substantially different potential that the metal you're touching in the first place. And if you are a parallel path, your own resistance is likely to be so much greater than the bonded metal parts that there isn't much of a hazzard. And then the overcurrent protection opens the circuit.

And Please can we some day dispel the myth that grounding (earthing) has anything to do with clearing a ground fault. The ground fault is cleared by the bond between the neutral and the bonded metals. This carries the current back to it's source with little resistance. The earth path has far too much resistance to motivate an overcurrent device. Drop a hot wire on concrete and tell me how many sparks you see...

I think the graphic meets the NEC.

Mike: Why in America do we require the neutral-ground bond at the main switchboard, why not simply allow the bond only at the Utility transformer in non dwelling-unit (commercial) applications? A stable 0V reference ground is essential to solid state circuit operation, such as amplifiers. Often pad mounted transformers have significant distance D from the main bonding jumper location, which is not ideal per figure 6-A. Jamie M. Fox, PE, LEED AP Mazzetti & Associates San Francisco, CA

The big problems are the "layers of civilization" in a building. It's the same as old minefields - undocumented, forgotten, ready to blow up. Companies will pay and pay and pay a little at a time and endure the high hidden costs of downtime and problems trying to duct tape the existing wiring, rather than just pay one big bill, gut it out right and do over clean & documented.

And I agree with Mr. Fox too. Why not just send in a 5-wire service?

I didn't see mention of the classic, when you have a noise issue, drill a hole in the concrete floor, pound in a stake and attach a ground wire and it will get all better. And other old wives tales...

MIke the figure 6.B is against the NEC???