Installing GFCI Breaker/finding the neutral wire

[Paulski]

I need to swap a normal breaker with a GFCI, problem is there is about 5 circuits coming from the same conduit and the neutrals are not labeled. So I need to figure out which is the correct neutral to connect to the gfci breaker, will this work:

-Turn off panel main breaker
-Discount all neutrals from the neutral bar
-Plug in a jumper wire across the hot and neutral on a outlet on the circuit
-Check for continuity at the panel from the hot wire to each neutral to find the correct neutral wire
-Disconnect the jumper wire and wire accordingly

 

[mcbane]

That sounds like a lot of effort, but it would work. If you have a volt meter and enough spare wire to reach from a known ground to the box, you can measure voltage from ground to each wire. Run a wire from the known ground to the box and use the voltmeter to test voltage vs ground. A neutral will be at or within 10 volts to ground, and a hot will be about 120 volts to ground.

 

[wyliesdiesels]

That sounds like a lot of effort, but it would work. If you have a volt meter and enough spare wire to reach from a known ground to the box, you can measure voltage from ground to each wire. Run a wire from the known ground to the box and use the voltmeter to test voltage vs ground. A neutral will be at or within 10 volts to ground, and a hot will be about 120 volts to ground.

if the circuit is fed from a main service panel, there should be no voltage potential between neutral and ground. even on a subpanel ive never seen voltage potential between neutral and ground since the main service panel bond is in place

a better way to trace this out is by shutting off all breakers for the 5 circuits, removing all neutrals from bus bar, add one back, then turn on the breaker for the circuit in question. then check for voltage between hot and neutral. Repeat the process until the correct neutral is found...

 

[Innovate1]

If you have a clamp on ammeter you could turn off the other circuits, put a load on this one and check the neutrals for current.

If not turn off the 5 circuits, disconnect the neutrals, plug in a load and turn on that circuit. Measure voltage at loose neutral wires. The one with full line voltage is the one. Faster than connecting them one at a time and testing each one.

 

[ycgoat]

There is more than one way to skin a cat. The OPs plan my have have difficulty because they may read continuity to the "hot" conductor through connected loads.
I would, similar to Wyllesdiesels, but with main power off; disconnect all neutrals in the conduit then at an outlet in the circuit in question check for continuity to ground at the neutral (with all neutrals disconected at the panel there should not be any). Then land the neutrals 1 at a time, checking for continuity at the outlet each time, until you get the one you are looking for. I like to connect it and disconnect a few times just to be sure. Wyllesdiesels's way is quicker and is all done at the panel but exposes you to an electrical shock and arc flash hazard, which should be left to qualified electrical workers.

 

[FredWanaker]

will a tone generator work in a situation like this with the 5 disconnected at the neutral bar?

Second question, if 5 neutral wires the same size go into a home run conduit from the panel, and all come out all together for 5 different circuits, why would it even matter which neutral is used? If one is replacing an existing receptacle would one not just use the neutral on it already? What am I missing here?

 

[ycgoat]

will a tone generator work in a situation like this?

They do make circuit tracers you plug into a power outlet, but you would still need to verify using another method.

 

[Innovate1]

Second question, if 5 neutral wires the same size go into a home run conduit from the panel, and all come out all together for 5 different circuits, why would it even matter which neutral is used? If one is replacing an existing receptacle would one not just use the neutral on it already? What am I missing here?

If they were using a GFCI receptacle that would be true. They are putting in a GFCI BREAKER.

 

[wyliesdiesels]

If you have a clamp on ammeter you could turn off the other circuits, put a load on this one and check the neutrals for current.

If not turn off the 5 circuits, disconnect the neutrals, plug in a load and turn on that circuit. Measure voltage at loose neutral wires. The one with full line voltage is the one. Faster than connecting them one at a time and testing each one.

thats a good way to fry a meter since youve now put it in series with the circuit... voltage should be measured in parallel not series

There is more than one way to skin a cat. The OPs plan my have have difficulty because they may read continuity to the "hot" conductor through connected loads.

I would, similar to Wyllesdiesels, but with main power off; disconnect all neutrals in the conduit then at an outlet in the circuit in question check for continuity to ground at the neutral (with all neutrals disconnected at the panel there should not be any). Then land the neutrals 1 at a time, checking for continuity at the outlet each time, until you get the one you are looking for. I like to connect it and disconnect a few times just to be sure. Wyllesdiesels's way is quicker and is all done at the panel but exposes you to an electrical shock and arc flash hazard, which should be left to qualified electrical workers.

only if touching live bus bars.... arc flash hazard is very very low on a residential single phase panel.

ive shorted hot and neutral together on a lamp cord and arc was very minimal....

will a tone generator work in a situation like this with the 5 disconnected at the neutral bar?

Second question, if 5 neutral wires the same size go into a home run conduit from the panel, and all come out all together for 5 different circuits, why would it even matter which neutral is used? If one is replacing an existing receptacle would one not just use the neutral on it already? What am I missing here?

you could use a tone generator, specifically one that changes tone when shorted, between neutral and hot as long as both are disconnected. connect tone generator at outlet then test with probe in panel. when you think youve found the right neutral wire touch hot and neutral together and tone should change. I do this for phone line work all the time

as to the shared neutrals, yes it will matter since the OP is putting in a GFCI breaker. connect wrong neutral and the breaker will trip. what youre missing is that the circuit neutral needs to be removed from the neutral bus and connected to the GFCI breaker.

 

[FredWanaker]

If they were using a GFCI receptacle that would be true. They are putting in a GFCI BREAKER.

got it. he needs to know what pair is coming from the receptacle, not going to a new GFCI receptacle.

1. Plug an incandescent light into the receptacle in question
2. Figure which breaker controls it, mark it
3. Turn off main breaker if possible for safety reasons
4 Turn off all breakers
5. Remove the five neutrals
6. Turn the main back on
7. Turn the one breaker back on
8. Look for voltage between the neutral bar and one of the neutral wires. Only one should have voltage - the one the incandescent is plugged into.
9. Turn the main off, install your new breaker
10. Turn on main and all breakers

FWIW I keep a 100' roll of flexible speaker wire around so I can use it with an ohm meter in situations like this. Turn the main off, unhook the neutrals, roll it out, hook one end to the neutral in question at the receptacle, and use the Ohm meter to find which one shows continuity. Done. Roll speaker wire back up.

 

[klassenl]

Honestly the clamp on meter is the probably the quickest and cleanest. I also understand that not everyone has a clamp on meter.

 

[sparky 1971]

There are several ways to do it and most of them have been mentioned. I'd just do the ol' continuity test myself. One thing that was left out: There are five circuits, but are there five neutrals? Four neutrals? Three neutrals? IF, the circuit you want to install the gfci breaker on happens to be a mwbc, that gfci breaker isn't going to work. Again, IF it's a mwbc, another neutral will have to be added specifically for that circuit.

 

[Paulski]

FWIW I keep a 100' roll of flexible speaker wire around so I can use it with an ohm meter in situations like this. Turn the main off, unhook the neutrals, roll it out, hook one end to the neutral in question at the receptacle, and use the Ohm meter to find which one shows continuity. Done. Roll speaker wire back up.

Wouldn't installing a jumper across the hot and neutral at a receptacle be doing the same thing in my case? The panel hot wire goes to the receptacle - the jumper connects it to the neutral wire at the receptacle - the neutral wire goes back to the panel, making one complete wire that can be verified by checking for continuity across the hot and neutral inside the panel. I can make a jumper with an old plug so I don't need to unscrew the receptacle from the wall

There is about 6 receptacles on this circuit, I don't know if they are interconnected using the lugs on the receptacle or with pig tails in each box, but I wouldn't think that would matter in completing the wire with a jumper and finding continuity at the panel

I'm thinking I will use this method to decide the first neutral to connect to the GFCI Breaker, then turn the power on and see if it was the correct one before connecting all the other neutrals back in to the terminal block. If its the wrong one, they switch the neutrals one at a time until I find the right one.

 

[FredWanaker]

I have an aversion to tying hot and neutral together in a troubleshooting test. It is is purely psychological. I hate smoke and sparks if I screw up somewhere.

 

[ycgoat]

Wouldn't installing a jumper across the hot and neutral at a receptacle be doing the same thing in my case? The panel hot wire goes to the receptacle - the jumper connects it to the neutral wire at the receptacle - the neutral wire goes back to the panel, making one complete wire that can be verified by checking for continuity across the hot and neutral inside the panel. I can make a jumper with an old plug so I don't need to unscrew the receptacle from the wall

There is about 6 receptacles on this circuit, I don't know if they are interconnected using the lugs on the receptacle or with pig tails in each box, but I wouldn't think that would matter in completing the wire with a jumper and finding continuity at the panel

I'm thinking I will use this method to decide the first neutral to connect to the GFCI Breaker, then turn the power on and see if it was the correct one before connecting all the other neutrals back in to the terminal block. If its the wrong one, they switch the neutrals one at a time until I find the right one.

The problem is on a 120v circuit the hot and neutral are tied together through the connected devices; it will work but you can also get false readings.

 

[Paulski]

I have an aversion to tying hot and neutral together in a troubleshooting test. It is is purely psychological. I hate smoke and sparks if I screw up somewhere.

I feel the same way. That's partly why I wanted to write it out to "disconnect the jumper" before turning power on, so I can use it as check list and not forget to do it.

 

[Innovate1]

thats a good way to fry a meter since youve now put it in series with the circuit... voltage should be measured in parallel not series

I said measure VOLTAGE on the disconnected neutrals. The meter isn't going to fry unless you try this with the meter set to measure current or resistance ( things other than voltage). I would be a lot more concerned with measuring resistance to find the wire. If for some reason there is voltage on the wires that could damage the meter. I always measure voltage to be sure it's zero if there is even a slim change there may be voltage before measuring resistance.

 

[mcbane]

if the circuit is fed from a main service panel, there should be no voltage potential between neutral and ground. even on a subpanel ive never seen voltage potential between neutral and ground since the main service panel bond is in place

a better way to trace this out is by shutting off all breakers for the 5 circuits, removing all neutrals from bus bar, add one back, then turn on the breaker for the circuit in question. then check for voltage between hot and neutral. Repeat the process until the correct neutral is found...

At the main panel that would be true, but he wouldnt be testing at the main panel. These are wires emerging from a conduit, and it is possible that other connections (and associated loads) may exist on one or more of the neutrals between the test location and the main panel. So depending what other loads any particular neutral may be carrying, there will in fact be a bit of differential voltage in the neutral.

While usually these differentials are small, I have seen 3+ volts differential on a number of occasions. Wouldnt expect to see much at a sub panel since the conductors are larger and loads on each ungrounded conductor are balanced to some extent.

 

[Paulski]

I am testing at the main panel, the wires come into the panel through a single conduit

 

[wyliesdiesels]

At the main panel that would be true, but he wouldnt be testing at the main panel. These are wires emerging from a conduit, and it is possible that other connections (and associated loads) may exist on one or more of the neutrals between the test location and the main panel. So depending what other loads any particular neutral may be carrying, there will in fact be a bit of differential voltage in the neutral.

While usually these differentials are small, I have seen 3+ volts differential on a number of occasions. Wouldnt expect to see much at a sub panel since the conductors are larger and loads on each ungrounded conductor are balanced to some extent.

actually he is testing at the main panel so there shouldnt be any differential...

 

[American Locomotive]

thats a good way to fry a meter since youve now put it in series with the circuit... voltage should be measured in parallel not series

You will never fry a meter measuring voltage in that manner. It's no different than putting a meter across hot and neutral. Voltage drop tests with meters are very common.

 

[FredWanaker]

I gave a safe 10 step process that would have taken 5 - 10 minutes. Someone else gave a simple clamp on meter solution that would have taken 2 minutes. Just plug an incandescent light in for the load on the receptacle and they will limit current somewhat. I know several people who keep an old light bulb in a holder around they can plug in for that reason. They even keep around an old 12V incandescent headlamp for troubleshooting 12V circuits for the same reason.