I'm going to list the 3 scenarios for electrocution I can think up for both a residential transformer with grounded neutral and an isolated transformer. I find that the grounded neutral transformer has one more scenario in which I get electrocuted than the isolated transformer does and I want to know why despite this, we reference neutral to ground in residential systems. I have heard it is for safety reasons but my thinking suggests that it is actually more dangerous to reference neutral to ground. I'm hoping someone can clear this up for me.

Imagine I'm standing at my main breaker looking at my hot and neutral lugs. For all of the examples below please imagine I'm not standing on any insulated platform. I'm barefoot on some kind of wet concrete floor. I'm also explicitly talking about the service entrance wires to do away with any complications coming from different branch circuits inside the home. Please assume I'm just talking about the service entrance wires.

I'm also going to talk about the current as if it were DC current and electrons come from the hot lug towards neutral. This is just because it helps get my ideas across in English easier. But please understand that I do realize that current will be flowing in both directions on an AC circuit.

Regular Residential Transformer w/ Grounded Neutral

  1. I touch the hot lug. There is a potential difference between the wet concrete floor and the hot lug. There is also an unlimited source of electrons (ground connection at neutral) connected to an unlimited acceptor of electrons (wet concrete floor). Current flows from the ground connection at the transformer, through my body, to the wet concrete floor and I get electrocuted.
  2. I touch the neutral lug. There is no potential difference between the neutral lug and the wet concrete floor since the neutral lug is grounded and I do not get zapped.
  3. I touch both the hot and neutral lugs at the same time. Current flows through my body (in addition to flowing through my body to the wet concrete floor) and I get electrocuted.

Isolated transformer (Neutral not grounded)

Now imagine that my residential transformer was not referenced to ground at the neutral point. Imagine it is like an isolated transformer.

  • I touch the hot lug. There is a (very likely - but unknown) potential difference between the concrete floor and the hot lug. However, there is no unlimited source of electrons like a grounded neutral. There is a limited source of electrons which is the wires inside the transformer themselves. I may get a small shock similar to static electricity shock as my body and concrete floor get brought up to the same potential as the hot lug but ultimately there can be no flow of electrons because there is no sustainable source for them to come from. They will push into my body because of the potential difference but at some point the transformer conductors will be so electron deficient that the force on the electrons to stay will be greater than the force they feel to push into my body.

  • I touch the neutral lug. It is unlikely that there is a potential difference between the neutral lug and the wet concrete floor so I will very likely feel no static-electricity like shock. In addition for the same reasons as above, even if there is a potential difference between the neutral lug and ground there is no sustainable source for electrons to come from so again I do not get electrocuted.

  • I touch both the hot and neutral lugs at the same time. Current flows through my body and I get electrocuted.

Electrocution Breakdown

  • Regular Residential Transformer w/ Grounded Neutral: Cases (1) and (3)
  • Isolated transformer (Neutral not grounded): Case (3) only.

So given this breakdown why do we ground neutral? Seems like it only introduces scenarios to get electrocuted. We could even have ground faults with an isolated transformer and not get electrocuted. Seems more dangerous to ground than to not.


2 Answers 2


I know you're looking for a "tweet" of an answer, a simplistic reason "Oh, it's this". There's actually a lot to it. It's not an ideology, it's hard empirical data culled from a sea of accident reports. They are using field data to "min-max" for minimum casualties.

What you're talking about is an isolated system.

That is a thing. And yes, if properly maintained, it amounts to "your first ground-fault is free".

I discuss the matter at length in this posting.

However, that "first ground fault" creates a treacherous situation for the rest of the system.

I had exactly that happen on a system I did not intend to be isolated. I found it when working on a circuit that I had turned off, and I flashed hot to ground just to be sure of it, and I got a tiny blue arc, and the fluorescent lights on that circuit came back on. The neutral was 120V from ground (whoopsie!) and the other phase was 240V from ground. Neutral was now dangerous, even though I had turned off the breaker, and it was a darned good thing I had done the "flash test" and chased down the problem before proceeding, because I was about to handle neutral in a grounded metal box, and that would've kicked my butt.

So "the first ground fault is free" is only clever if you have maintenance people actively searching and testing for that first ground fault, and they repair it BEFORE a second ground-fault occurs. Otherwise, it simply fails in silence, creating a time-bomb that is a worse situation than a grounded system, and bites you later with very unexpected behavior.

Indeed, Code does allow isolated systems on sites where on-site staff do active maintenance and monitoring of the system. That's sure as heck not a house.

  • 2
    And further: small leakage currents to ground are extremely common in large installations. Otherwise we could just put GFCI devices near distribution transformers, but there is enough leakage currents that it would trip all the time - this necessitates installation closer to load, to keep the total leakage below 30mA. Similarly an isolated neutral would float at a dangerous voltage even in a non-faulty installation, just because the total leakage in a block of houses is too large.
    – jpa
    Nov 21, 2020 at 8:25
  • 1
    Thanks @Harper/Monica :) If I were to distill your answer down to "a tweet" :P here is my basic takeaway: In a center-tapped residential transformer, not bonding neutral to ground means that in the event of a ground-fault, the potential differences of each service wire wrt ground change. Neutral goes from 0v to 120v, the ground-faulted hot goes from 120v to 0v, and the other hot goes to 240v. The unreliability of these voltages wrt ground is in itself problematic and as your anecdote illustrates - can be dangerous. Bonding neutral to ground provides reliability of voltages. Is this correct? Nov 23, 2020 at 20:31
  • @natsuki_2002 Yes, that is one way it works in your favor. Nov 23, 2020 at 21:47
  • 2
    In an isolated system "your first ground-fault is free". And the second one can kill you, with no further warning. In a grounded system the first ground-fault should blow a fuse/trip an MCB and safely cut off the power.
    – Graham Nye
    Nov 26, 2020 at 0:30

Neutral is grounded in power distribution systems because power lines are on the top of poles and are enticing places for lightning strikes. It is not grounded to enhance electrocution safety for people who want to install electrical outlets in their shower stalls to power their radios while they take a shower...LOL

By grounding neutral they can install spark gaps in the distribution network to allow lightening to be bled off to ground instead of coming into people homes over the downlead and exploding their TV sets off the wall.....

If you had an entirely underground distribution system you could forgo the ground on the last stepdown into the customer premise and probably make it safer.

I hope you get an A on your essay... ;-)

  • 1
    Thanks Ted :) So sounds like lightning protection is the biggest reason that neutral is grounded. I had read somewhere that people living in homes that have an accidentally "floated" neutral experience things like "tingly faucets". How does this play in? Nov 20, 2020 at 20:32
  • It might be possible to induce a stray current but more likely, they have a panel that is grounded through a metal cold water pipe - and maybe the water feed from the street meter is replaced with PVC - leaving some metal parts of the plumbing bonded to the neutral in the panel but not to a metal water pipe going into the soil. In some places with acid soil they may be going to plastic for the feed pipes. I'm sure a lot of plastic feed pipes were put into Flint, MI after the previous governor attempted to poison the entire town's water supply.....LOL.. Nov 20, 2020 at 21:08

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.