I have done a lot of basic house wiring, but have stubbed my toe on something new to me. I was installing a light fixture in a branch circuit at the point closest to the breaker, and opened a splice in the neutral, isolating the neutral side of the circuit from the panel. I got a little zap, and discovered that there was a voltage difference of 19-27 volts between the isolated neutral and the grounding wire. The building was vacant with NO loads on the circuit. I disassembled every outlet, splice & switch on the circuit, and the voltage from neutral to ground varied from 19 to 27 during this process. What is going on?
Discussion Forum
Discussion Forum
Up Next
Video Shorts
Featured Story
Listeners write in about saving frogs, attic ductwork, lime wash, and energy ratings for real estate.
Featured Video
How to Install Cable Rail Around Wood-Post CornersHighlights
"I have learned so much thanks to the searchable articles on the FHB website. I can confidently say that I expect to be a life-long subscriber." - M.K.
Replies
First of all how are you measuring it?
With a digital or high impedence VOm you will often get a measure on signal capacitively coupled to the open wiring. Ie, the wire is acting as an antenna.
But there is not enough energy that to cause a shock. But I have found that sometime I have bit of metal sliver in my finger or a stuck with a bit of fine stranded wire and it feels like a shock. Then if you use a high impedence meter you see the stray voltages and thing that is what you felt.
If there was truly NO LOAD (don't forget lights and things like a doorbell transformer) and you measure the voltage with a wiggler or while you have a small load on the circuit then you have insulation breakdown.
"...isolating the neutral side of the circuit from the panel...."
Is the neutral grounded anywhere else, even a grounded tool laying on another ground? The capacitive leakage to read 20 volts on a 10 megohm meter would be about 25 - 30 pF, which is about right for what you are reading. On the other hand, getting a zap strong enough to feel would need about a mA, which would need about a mile of romex to get a 50 nF capacitance required to allow that magnitude of current.
1. Put a lightbulb across the neutral gap and measure the voltage.
2. if the bulb glows even very faintly, your isolated neutral is NOT isolated - could be from multiple reasons, including a partial insulation failure .
3. Even with the kW-hr meter on the outside of the house, you can get neutrals to spark if the 'isolated' neutral is not isolated and the ground currents from the neighborhood share thru that path.
You need to determine for sure that you do not have a partial insulation failure somewhere. If the voltage is near zero in #1 above, you are OK there, and probably the major problem is with you neighbors (people on the same distribution transformer as you are) grounds.
Hartman is correct. The method used to measure voltage makes a big difference. Most digital multimeters are high impedance units that are quite accurate reading very small voltages but they are so sensitive and so poor a drain that they are prone to ghost voltages that are of no consequence in most line voltage wiring.
A low impedance unit, the most popular being a "Wiggy" solenoid voltage checker, is less accurate, can't effectively differentiate between voltages of less than about 60v. On the up side a low impedance unit drains any stray induced voltages so you don't go chasing ghosts.
Just to be safe have someone experienced look at your neutral/ground connection at the first point of disconnect. Check the neutral there and in any subpanels. A weak neutral connection could cause this symptom. I don't want to panic you and this is probably not the case but a bad neutral can cause major problems so get it checked ASAP.
No reason to camp out in the snow but have it looked at in a day or two.
You can also turn your high impedance meter into a low impedance meter simply by putting a resistance in parallel with it. I made one by attaching a power resistor to one of those dual banana plug things. It plugs into the meter, and then I plug the test leads into it. As a temporary rig, you could use an ordinary incandescent light in parallel with the meter.
-- J.S.
Can't say that I have ever done, or even thought of, doing that but with what little I understand of the intricacies of theory it should work. I'm seldom without my Wiggy but that's a nice tip. If I get a chance I'll run the numbers on the technique. Just to help me remember it.
Had this same kind of problem about a month ago at a friends house. After chasing ghosts for about an hour I decided to check the breaker. (you never check the easiest things first, do you). After killing power to the panel, I disconnected this circuit lead and swapped it with another circuit on another breaker and low and behold no more AC trickle. After coming to the conclusion that the breaker was "bad" I told the HO friend to get a new breaker and he'd be back in business. The next day while at a party at the HO's house I was talking to an HVAC friend and he said there was nothing wrong with the breaker. the connectors just needed to be pinched tighter with a pair of pliers to restore the connection. He was right. The breaker wasn't "bad" it just had a bad connection.
So like 4lorn1 said, it could be a faulty neutral connection. That's what it was in my case; it showed the same symptoms as you described.
Dan
Took your suggestion. Tightened neutrals at the bussbar in the service entrance, and the grounding clamp (on a water pipe). Wow. Now I get 120 volts between the "Isolated" neutral and the grounding wire.
Either you have an open neutral and you are seeing 120 volts "backfeed" from a load that is between a hot and the open neutral or the you have a hot and neutral reversed someplace.
Or this "ground" might be hot.
Go back to the main disconnected and check with the meter there and verify that it is correct.
Then start working down the line.
And you might want to get a long wire that you can connect back at the a known ground at the main disconnect and use that as a reference as you work down the circuits.
Hmm. tightening the connections didn't solve the problem but it did simplify it. A bit of progress.
I agree with Bill Hartman. Starting at the first point of disconnect, directly after the meter, work your way back to find the fault. If you get a bad reading, at the first point of disconnect where there is a common connection for both the neutral and ground if this doesn't read near 120v, 115 to 125v, to either hot, just to be sure try reading it with the main breaker off also to eliminate the possibility of a back feed, just after the meter call the utility. A bad neutral connection at the meter or on their lines is possible.