Electrical-28 volts through ground

Page 2 of 2
• posted on February 21, 2006, 12:35 am
Very good response. At least for me, as I was actually able to understand it.
However, I do have a followup question. I recently replaced a fire alarm system in a condo tower which had the amplifier stack, fire alarm system, and security panel on 3 seperate hots but used one neutral.
I opened the disconnect for the ampifier stack, and fire alarm panel, and measured from hot to neutral (0V), Hot to ground (0 volts), neutral to ground (0v). When my assistant opened the neutral he noticed some arcing as the security system was still live. Thus I learned about the multi-wire branched circuits.
How would i go about knowing if a neutral is live or not? As all my measurements above, and (gasp, I know it's not a good method) my magic wand fluke voltage checker all showed 0V.
Thanks, JW

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 21, 2006, 6:02 am

You can't, really. The best you could do is use some sort of current detector (eg: clamp-on ammeter) to look for current flow in the neutral before you disconnect it. Alarm systems may be carrying quite low current, and a clampon ammeter might not register that low.
"Shared neutral" is usually used to refer to a 4 wire 240V circuit, where you share the neutral between the two hots. Or a 5 wire 3 phase circuit, where you share the neutral between the three hots. In these circuits, the neutral is only carrying the _difference_ between the hot currents.
When you share neutral between two hots on the same leg of the panel, it's possible to overload the neutral, because the neutral current is the sum of both hots, not the difference.
The two paragraphs above are because I _suspect_ that you're not really seeing a shared neutral - especially because of the reference to three hots - unless the alarm system is hanging off a 3 phase circuit.
What I more suspect that it's actually a _single_ 120V circuit, you disconnected an intermediate junction, and the "disconnect" only disconnected _part_ of the circuit, whereas the neutral fed thru to more equipment that got its feed _before_ the disconnect. Eg: the downstream equipment "feeds through" where you disconnected, but the downstream hot didn't.
Mind you, I've not worked with large multi-unit buildings, so maybe they are wired differently.
Shared neutral circuits are supposed to be wired such that you can remove a device in the middle of the circuit without interrupting the neutral or ground. Eg: pigtails.
If you're working on a suspected shared neutral circuit, when you have the slightest doubt, you should take the time necessary to track down where all the hots are, and kill them all at the panel. I personally will not wire shared neutral circuits without a tied breaker, despite the fact that code doesn't always require it.
--
Chris Lewis, Una confibula non set est
It\'s not just anyone who gets a Starship Cruiser class named after them.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 20, 2006, 3:11 pm

Yes. But as I mentioned elsewhere, phantom voltage or not, measuring voltage between ground and neutral means there's a very real problem.
Grabowski mentions the possibility of a loose neutral in a shared neutral circuit. That indeed is possible. However, I think it more likely that somewhere upstream of that outlet there's a poor ground connection. Many not-quite electricians are quite sloppy at connecting grounds together.
A loose neutral in a shared neutral circuit has other side effects. Like having lights brighten when you turn something else on. The circuit will seem "sick" in other ways. Besides, shared neutral circuits are usually only a tiny fraction of the circuits in a home.
--
Chris Lewis, Una confibula non set est
It\'s not just anyone who gets a Starship Cruiser class named after them.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 20, 2006, 5:56 pm
Chris Lewis wrote:

Disappear? Not necessarily. On a 150 volt scale a 20,000 ohm per volt meter has a resistance of 3,000,000 ohms. A hot-ground 'leak' of 12 meg ohms would give reading of 28 volts from 'ground' to ground!.
bud--

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 20, 2006, 8:38 pm
wrote:

OK, you did say that.

OK. I'll abandon that idea. Thanks

Remove NOPSAM to email me. Please let me know if you have posted also.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 19, 2006, 8:37 pm
Pop wrote:

On the safety ground? That seems pretty unlikely to me. The safety ground should have a low impedance path to earth.

--
The e-mail address in our reply-to line is reversed in an attempt to
minimize spam. Our true address is of the form snipped-for-privacy@prodigy.net.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 19, 2006, 9:48 pm

Yes, phantom voltage can be puzzling to some. I can remember my first high impedence VTVM, I found that it could read voltage between my left and and my right hand. I guess this is one of the principals of a lie detector.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 20, 2006, 12:45 pm

the
A few things I would do: Check the voltage between that suspect ground wire and the nearest water pipe. Also check the voltage between the other wires in the switch box and the water pipe. Next I would shut each circuit breaker off in the panel one at a time and check to see if the voltage reading changes.
I would also go to a different part of the house on a different circuit and check the voltage in the same manner with the circuit off.
Let us know what you find.
How many wires are in the switch box?
I'm thinking that this particular circuit may be part of a three wire circuit. The voltage that you are reading is actually coming from the neutral because there is a load on the other leg. By shutting off the other circuit breaker the voltage should almost disappear. If this is the case I recommend that you tighten all of the neutral and ground connections in the electrical panel. There could also be a neutral connection somewhere on the circuit that needs tightening such as a receptacle.
John Grabowski http://www.mrelectrician.tv