While I was replacing an old outlet, I received a shock from something in the box. I think it was the metal box itself.

Before I started, I performed the following to ensure the power was out:

  • Plugged a lamp into the outlet and turned it on
  • Flipped the breaker off; verified that the light went out.

I then started to disconnect the wires when I received a shock.

After I flipped the breaker for the whole house, I replaced the outlet without any further problems.

Should I be concerned that my ground isn't really grounded?

  • 5
    Maybe there's more than one circuit in the box?
    – Steven
    Commented Jan 27, 2012 at 1:35
  • How old is the house and/or its wiring?
    – Skaperen
    Commented Jan 27, 2012 at 2:49
  • The house is mid-1970s as is its wiring.
    – kmm
    Commented Jan 27, 2012 at 4:24
  • Pictures might help. Also knowing how many wires are in the box, and if any are there that are not connected to the receptacle would help.
    – Tester101
    Commented Jan 27, 2012 at 13:00

11 Answers 11


Well, if there is something leaking current onto "ground" and that thing is connected to your metal box, and the metal box is not connected to the ground in your breaker panel, no, that's not great. I would take a multimeter, turn off the breaker to the house, and check the resistance between neutral and the box. If it is large, then they are not not bonded together at the panel, but yet something somewhere is connecting them in your house. Try using a 3-prong tester on some outlets in your house and see what results you get.

  • 6
    Note that 3-prong testers can't really tell you that ground & neutral are wired correctly, for example if they are swapped.
    – Jay Bazuzi
    Commented Jan 27, 2012 at 1:21

This is an old thread, but it caught my eye. I experienced a shock when the breaker was off while working inside a box for a light switch. After a lot of head scratching, I realized the lights were all fluorescent floods that contain a capacitor for faster startup! Discharged the capacitor to ground and no more voltage!

  • 2
    Capacitors will get you if you don't watch out! Commented Dec 4, 2015 at 0:16
  • I've seen caps running at less than 3V throw visible sparks when shorted. (This was on a PC motherboard, btw.) Commented Dec 4, 2015 at 0:55
  • Just ran across this. Needs more love. Commented Apr 27, 2016 at 18:06

A ground not really connected to ground will not, by itself, create a shocking situation. A proper ground could, depending on the cause, complete a short circuit that would be tripped a breaker in the past when the real cause was established.

You may simply have two circuits in the same box. For example, a shared-neutral circuit would be operating from two separate breakers on opposite phases. The receptacle may be a split type (the breakable tab between the screws is broken off). Or another circuit may be passing through (such as alternating circuits between outlets).

You need a voltage sensor device to begin with. This situation is one of many reasons why smart electricians use these. The proper use would have detected a "hot situation" and allowed you to avoid being shocked. It works better than a voltmeter because it avoids having to move and contact the wires.

Once a voltage sensor is showing the situation, it can aid in resolving it as well. First, unplug all other appliances in the house to see if any of them are leaking voltage back in to the circuit in question. This may happen from a pair of interconnected devices on different circuits.

If unplugging all devices from the circuit does not resolve it, the next step is to determine which other circuit is feeding the voltage, to narrow down what might be cross wired or shared. You can do this by switching off (or on as the case may be) half the circuits at a time. Once you know a set that is causing trouble, work them in half, repeat until narrowed down to one.


If you're sure it's not another live circuit in the box, the next most probable cause in a house built in the '70s is a neutral-to-ground short or swap. I personally think it's unlikely to be a hot-to-ground leak, because you'd be blowing fuses left and right, including when you get your electric bill.

In houses originally built without three-prong grounding, "poor-man" retrofits were often done by hooking the ground to the neutral. This isn't a code-compliant path to ground as it doesn't provide shock-safety, but since the neutral of a circuit should always remain closed, it's at least an always-available path that's hopefully easier than through you, lessening the severity of a shock, and if a full hot-to-ground short should occur the fuse/breaker will still cut out.

Additionally, someone who didn't know what they were doing when they wired a plug in your home could have swapped the neutral wire for the ground wire. This is unfortunately rather common, especially in homes that have been retrofit to add grounding (the ground wire can't be bare and thus easily identifiable if it's not inside the Romex bundle, so someone could mistakenly hook up the green-jacketed wire as neutral, or could conceivably be faced with a box that has two indistinguishable white wires). when something is plugged in to that outlet and turned on, any other ground wire that is more directly connected to this outlet than to ground, or which is along a series path to ground, will be energized.

Both of these will pass a home inspector's basic tests with a three-prong tester; the average tester cannot tell the difference between neutral and ground, as they are both connected to the same bus strip on the service panel (which, confusingly enough, has further connections to both the common wire going back to the power company, and to your home's plumbing aka earth). A multimeter can help detect N-G "shorts" (there will be some resistance between neutral and ground normally, but virtually none if the neutral was shorted to ground within that box), but really the only way to be sure your home's wiring is safe is to open up every outlet and switch in your home and ensure they're wired correctly.

  • But neutral and ground are bonded at the main panel, so won't there be continuity?
    – Steven
    Commented Jan 27, 2012 at 20:44
  • There will be, yes. I'll edit. I had thought that the ground went to the neutral bus THROUGH earth, not to the panel and then to earth. The way I thought of it would introduce high resistance between the neutral and ground, as both would have more affinity for earth than each other.
    – KeithS
    Commented Jan 27, 2012 at 21:31

One scenario under which it is possible is the breaker you flipped was single-pole and disconnected the neutral wire instead of the phase wire because of wiring done incorrectly. With neutral disconnected the light would of course go off, but touching a life phase wire alone can get you shocked and sometimes depending on the situation that can be fatal.

  • A "hot-neutral swap" is possible, but unlikely as the breaker itself had been cut at that time. When the breaker is off there should be NO power coming in on that circuit. What's more likely is that there's something else miswired close enough to this box that it's putting potential on the ground side.
    – KeithS
    Commented Jan 27, 2012 at 19:25

The only way this could happen is if there is no ground wire, or it is connected to something it shouldn't.

It could be that the ground was never connected at the main panel, and the ground of one of the outlets touched another wire. It doesn't matter whether it is touching hot or neutral. ( Neutral isn't quite 0 volts at the end of a line, especially if there are inductive loads.)

It's also possible that the cable runs next to another cable that is still live. The live wire can induce significant voltage into a wire that isn't grounded.

The only way to be sure there isn't any AC voltage, is to use a non-contact voltage sensor.

I would recommend calling someone who knows what they are doing, and have them take a look.

  • 1
    Any induced voltage from cable proximity (phantom voltage) would be very low amperage, and would likely not result in a shock.
    – Tester101
    Commented Jan 27, 2012 at 14:13

I was replacing an outlet and had the breaker turned off. There was still a small amount of current through the outlet. I had an electrician check it out and it turned out to be from a night light plugged into an outlet in another part of the house. He unplugged the night light, end of problem.

  • 2
    That sounds more like the beginning of the problem than the end -- what wiring problem allowed a night light plugged into one outlet to cause voltage to be present on a different outlet on a different circuit?
    – Johnny
    Commented Jan 14, 2015 at 0:14
  • 1
    @Johnny Half of a MWBC still on (with a load), while you're part of the neutral = ouch.
    – Mazura
    Commented Mar 5, 2015 at 1:20

A MWBC (multi-wire branch circuit, where two hots share a neutral) incorrectly wired to two single pole breakers, instead of properly using a 2-pole (tied handle) breaker, will only allow you to kill half of that Edison circuit by flipping one switch. You'll have to find its sister and kill that one too. If not, once you disconnect that neutral you can (and will) become the path of least resistance if there is a load on that side of the branch. Just like Jere's night light; the reason why it was 'safer' once the load was removed. -I'd bet money his was a MWBC.

If a neutral sparks when disconnected, stop and find that other breaker.


You could have cross polarity in the main board where the breaker is off but the neutral bar is feeding the board backwards. That's why you always test with a probe tester from live to earth even when you know you have switched it off.

Professionals always do this because sometime there's no coming back from a good shock!


Someone actually wired a 3 way switch in my house using the traveller screw as a ground. It took forever to chase down do to the fact it was an intermittent problem that only happened when the light was off.


Possible double-fed. This happens ( moreso than it should ) when two breakers of apparently different branch circuits, but same phase, accidentally, or incorrectly become tied back somewhere downstream.

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.