4
\$\begingroup\$

Continuing my attempt to understand the use of grounds in modern electrical service: There are numerous posts illustrating why a "bootleg ground" (ground connected to neutral outside of an AC service panel) is dangerous. But is it fundamentally more dangerous than other similar wiring faults that can occur?

By code, I gather that the introduction of line voltage onto the third "ground" line in an a electrical service is always the result of a wiring fault. It's unclear to me that the ground service loop has improved safety. I can see examples where it averts a danger, but doesn't it also introduce hazards where others did not previously exist? For example:

  1. Single-device fault: In a two-conductor service, without the ground, a device wiring fault could either expose users to line voltage (leading to possible death) or else short the circuit, causing the circuit breaker to cut power. If the device is grounded then the device can be constructed with a "grounded case" which, in the same fault, attempts to catch and divert the line voltage to the ground so the user doesn't contact it.

  2. Ground wiring fault: If you put a grounded device with a ground fault on a circuit with a broken ground, now all grounded devices are carrying line voltage on the ground. In fact, because device grounds are designed to do their best to isolate users from voltage, they tend to connect to cases and enclosures users are likely to contact.

So it seems like adding grounding to an electrical service can possibly avert a hazard in device fault scenarios, but at the expense of introducing a new hazard in the event that occurs together with a service ground fault.

Now, I suspect I'm missing important elements of this analysis, since grounded service has been code for a long time. Can someone fill in the blanks to illuminate why grounding AC service is clearly better then two-conductor?

\$\endgroup\$
6
  • \$\begingroup\$ Whose code? Location is likely to make a difference here... \$\endgroup\$
    – Solar Mike
    Commented Jul 27, 2018 at 20:25
  • \$\begingroup\$ @SolarMike – I'm not asking specifically about code, but rather just noting the prevalence of the ground wire in electric codes as a hint that it is considered a "clearly good thing." So if you want to reference a code pick any one that specifies a ground wire for electrical service. \$\endgroup\$
    – feetwet
    Commented Jul 27, 2018 at 21:18
  • 1
    \$\begingroup\$ But codes vary - is the UK code the same as the Swiss or as the USA NEC? So, the answers are likely to be different - and I know that the codes for UK and CH differ... \$\endgroup\$
    – Solar Mike
    Commented Jul 27, 2018 at 21:21
  • 1
    \$\begingroup\$ @SolarMike: If you can find a current electrical code for AC service anywhere in the world that specifies a ground wire and allows for it to be "boot-legged" to the neutral at any point short of the primary service panel, then that would be worth noting in an answer. I am assuming no such code exists. \$\endgroup\$
    – feetwet
    Commented Jul 28, 2018 at 2:44
  • 1
    \$\begingroup\$ More details on the risks of bootleg grounds are over at DIY.SE, e.g. diy.stackexchange.com/q/97390/2196 \$\endgroup\$
    – BMitch
    Commented Jul 30, 2018 at 12:42

2 Answers 2

4
\$\begingroup\$

Electrician here. Answering in terms of the Canadian Electrical Code, and also my personal safety as an electrician interacting with things that have been worked on by others. I'm going to use the word "Ground" here in the colloquial manner, but technically in electrical, the system itself, where it physically attaches to the earth is grounded. In all other locations within the system, it is considered to be bonded, so you bond the whole system/equipment/building/together, and separate from that, you ground it.

As for the bootleg ground, the simplest answer is that anything at all that causes the system, particularly the safety elements not to function in the intended manner can lead to loss of life. As for whether it is more fundamentally dangerous, some of the time it will be, some of the time it won't. Even with a 120v 15A circuit though, it could mean your life. The most fundamentally dangerous things in electrical are things that you would not expect to either run 50-100mA AC through your heart (which 120V it turns out is just right for for most people) or that can run enough current through you to ensure rapid disassembly. Vastly different amounts of current, but both, ideally, require the element of being something you would not expect.

1.) Single device fault: The intent of the ground wire in textbook terms, is to provide fault currents with a low impedance path to ground.

Inside of buildings, this ensures that should a wire be shorted to ground, the surge of current will be sufficient to immediately blow the breaker (although some brands have batches of breakers that you can literally weld off of while they hum away and fail to pop). On the outside of the buildings, the electrical company is actually depending on every house having an adequate attachment to ground, as this accounts for a portion of the grounding for their entire distribution system.

Inadequate path to ground on a massive (by electrician standards, not distribution standards) distribution voltage breaker can lead to incredibly large failures, dangers and problems.

As for the "grounded case", if it fails to provide an adequately low impedance to ground, it is inadequately designed, and we won't see it commercially in Canada due to it's inability to pass CSA or UL testing. We are still able to order some fairly dangerous objects directly from other countries if we don't know what we're doing, but in order to use them in a commercial or insured product/project, you would need to get an appropriate agency to sign off on them. All of this said, case grounding is well understood and very far from the more difficult aspects of engineering. In countries with better safety codes (Our country looks pretty unfavorably on unnecessary worker deaths and our codes reflect that) every permanent metal object save fasteners etc, and certainly all that would be expected to come into contact with a live conductor will be thoroughly grounded, so long as no one messes up.

2.) Ground wiring fault: To the degree that the situation you are describing can exist, yes, having energized objects that people can touch does present a danger(as a first year apprentice I had to stop an out of country exchange program worker from repeatedly putting his trowel in wet 347V mud. The mud was near a post that had been energized to 347V by an unlucky drilling of a large concrete insert. They had drilled into a lighting conduit for the floor below and either not noticed or they were too embarrassed to report their error to their foreman, so the friendly tile worker showed me by repeatedly shocking himself.)

Freak incidents do occur. If that post had been grounded though, as soon as that lighting circuit had been turned on, the breaker would have popped, removing any danger and revealing that something needed to be troubleshot.

As for the broken ground, the way modern buildings are constructed it is actually surprisingly difficult to achieve in the first place for anything other than small wood frame buildings. In a small wood building it's theoretically possible, but the parts of the connection that would need to be broken are protected by a wall, handled by professionals, and tested before handed off for commercial purposes. This means that while during the construction of a building I as an electrician have to worry about this, as do coworkers from other trades, the end consumer never should, so long as they adequately maintain their house.

The worst I've seen in a building handed over to the customer was probably a broken neutral (also very dangerous) in the building I currently live in, done by who knows who, but likely not another electrician. Second worst was a light switch (not light or ballast) that the owner said buzzed whenever it was on. Someone had not adequately tightened a screw, causing it to constantly arc. They had, however stripped a full inch of wire, bent it around the screw and pinched it with strippers, and that likely prevented something worse from happening.

One other thing is that some safety devices, like GFCIs you find near water or liquids, rely on the ground to function. They're great. They're extremely sensitive, and all they do, is if every bit of current going out on the line doesn't come back on the neutral, they break the circuit. If you get shocked through a functioning gfci outlet or breaker, you will never get locked up, I don't believe enough current can run to stop your heart, and you may not notice the shock.

Lightning arrest systems also rely on grounding systems.

It would be more fair to say that adding grounding to an electrical service certainly averts a vast assortment of hazards, both related to the system and devices connected to it, at the expense of increasing necessary materials, training and engineering costs, and poor installation leading to flaws in the system can still create hazards, in the same way that poor installation could create hazards in a groundless system. If someone finds a way to "acceptably" install something that creates an unwarranted hazard, we upgrade our electrical code. Simple as that.

The fact that it takes some time to think up a scenario in which a grounded system is more dangerous at all than an ungrounded one says something. I could probably come up with a few too for conversation, but I'd rather spend the effort elsewhere and I know whatever I come up with won't make me any less grateful for a grounding system.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Thanks; this is an illuminating answer. Nitpick: GFCIs don't require a ground to function. (Though perhaps they do to implement a convenient "test" function?) \$\endgroup\$
    – feetwet
    Commented Jul 28, 2018 at 2:51
  • 1
    \$\begingroup\$ GFCIs do require a ground to function, not necessarily a ground wire, but there must be some path for the current to return to source other than on the neutral, which while it may not be a designated bond or ground wire or material, is almost certainly the earth. In this case, the ground connection in the form of rod, copper plate or buried wire connected to the system at the panel is crucial and while the actuator does not rely on the ground to function, it does rely on every bonding connection in the system to be maximally safe. \$\endgroup\$
    – K H
    Commented Jul 28, 2018 at 3:07
  • 1
    \$\begingroup\$ Take the modern toaster in the bath scenario. The worst thing that could somehow happen would be for your connection to ground in the bathtub to be so poor that the neutral took all of the return current, which would cause it to continue conducting. This worst case scenario is an example of why GFCIs have to be so sensitive. \$\endgroup\$
    – K H
    Commented Jul 28, 2018 at 3:10
  • 2
    \$\begingroup\$ @KH - I think that feetwet means that the GFCI breaker itself doesn't require a ground. Hot and neutral flow through a transformer, and if the two currents don't exactly match, you get a trip signal. A ground connection elsewhere is needed to divert some current, but not at the GFCI itself. \$\endgroup\$ Commented Jul 28, 2018 at 14:23
  • \$\begingroup\$ Definitely true, covered in my first comment. The GFCI device does not need the ground wire to physically actuate, but it's performance in emergency situations is improved by every bond in the system, and as a result it relies on the grounding/bonding system to be safe. \$\endgroup\$
    – K H
    Commented Jul 28, 2018 at 19:06
4
\$\begingroup\$

If the neutral breaks a bootleg ground is deadly.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Right, but this begs the original question: the alternative is "the good old days" in which there was no ground available to a device, so a broken service neutral couldn't energize the case (unless there was also a device fault that causes the line to contact the case). The question is why, despite the introduction of this new hazard, grounding is considered a clear net benefit. \$\endgroup\$
    – feetwet
    Commented Jul 28, 2018 at 13:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.