Strange circuit breaker activity?

[hanish123]

Hi there. I'm dealing with a difficult to locate problem with our 120V landscape lights. My first thought was there was a short somewhere in the run, but there seems to be something additional going on at the circuit breaker or closer to it. The circuit (one bundle of white, black, green) goes from the circuit breaker on the side of the garage to a junction box also on the side of the garage, where it connects to two different bundles, one serving the front landscape lights and the other serving the back. When I disconnect all of the wires at the junction box and turn on the circuit breaker, I get voltage on the black wire from the circuit breaker and not on any other wires (which is correct). But when I hook up the wires from the circuit breaker to either of the bundles in the junction box (i.e., either the front landscape lights or the rear landscape lights), then there is no voltage detected on any of the wires in the junction box, even on the black wire coming from the circuit breaker. The circuit breaker does not trip, but somehow the voltage is gone. If I disconnect the landscape lights wires from the circuit breaker wires again, the voltage comes back on the black wire coming from the circuit breaker, again without resetting the breaker or anything. It's like the circuit breaker is cutting the power on its own without tripping, or perhaps something else that I'm not knowledgable to know. Any ideas? The circuit breaker does seem to work by tripping if i introduce a short on the line (touch the black wire from the circuit breaker to the white wire from the circuit breaker), but I'm just not sure what could cause power to essentially be cut off to a circuit merely by attaching a black wire to any other black wire without a circuit breaker or GFCI being involved.

 

[westward10]

Moved from DIY.

 

[westward10]

What are you measuring this voltage with. Introducing a dead short between hot and neutral really isn't the greatest of ideas.

 

[Lucien Nunes]

then there is no voltage detected on any of the wires in the junction box, even on the black wire coming from the circuit breaker.

When you make these voltage tests, what device are you using and where are you connecting its other probe? I.e. the voltages you are measuring are being measured with respect to what reference point?

The behaviour is normal for a high resistance connection, other than the bit about making the breaker trip. A voltmeter uses so little current that enough will get through a nearly broken connection to give a full 120V reading. But when the lights are connected, the tiny current able to get through disappears into the lights without significantly raising the voltage, and the meter reads near zero. In UK parlance we call it a 'ghost voltage.'

Normally, that would also prevent the breaker tripping even when short-circuited (as above I agree this is a bad idea!). So we need to understand what tests you are doing to understand this anomaly. It is possible that the high resistance is within the breaker itself, and when the short-circuit is applied it arcs and momentarily conducts well enough to trip, but I would still expect the lights to work / flicker and the ghost voltage not to collapse completely.

 

[hanish123]

I'm just using a non-contact voltage sensor. I have a multimeter, but not sure if its really necessary to use as my problem is so dramatic (the power is completely cut off in the circuit or completely on but doesn't appear to be in between), but I'm open to suggestions. The fact that neither the front or rear landscape lights work, even when individually attached to the live wire, suggests to me that there must be some problem in the circuit breaker itself or in the wires from the circuit breaker to the junction box, no? Unless both subcircuits happen to fail at exactly the same time. I just can't figure out what that problem would be. I've certainly never seen this type of behavior in any of the other circuits I've worked on or run, when testing using the non-voltage contact sensor. I mean, if I open up one of the house outlets and put the non-voltage contact sensor on the black wire the sensor will sound whether that black wire is connected to the receptacle or not, assuming the circuit is live, but are you saying that lights work differently?

 

[westward10]

A non-contact voltage tester although handy in some circumstances is of no help here. You need an approved contact voltage tester, not a multi-meter.

 

[brianmoooore]

What are you measuring this voltage with. Introducing a dead short between hot and neutral really isn't the greatest of ideas.

And there was you lot saying, the other day, that you couldn't test second hand ebay breakers.

 

[Lucien Nunes]

You appear to have a high-resistance contact or connection either at the breaker or wiring to the junction box. A non-contact voltage sensor is not ideal for tracing that fault as it cannot tell the difference between a wire that is properly live, and one that is live only through a high resistance. However, what you can do is to reconnect the lights, switch the breaker on, prove that there is no voltage at the junction box and then check whether the black / hot wire is live where it leaves the panel. If it is, the bad connection is in the run of cable. If it is not, the problem is likely the breaker itself or faulty contact with the busbar. What model of panel is it?

A better method is to isolate the power at the panel and trace the break using the resistance range on the multimeter. This will actually locate the bad connection, rather than merely the symptom it produces. It is also safer to work with the power off than on.

You need an approved contact voltage tester, not a multi-meter.

Although, as they don't work to GS38 in the USA, these seem to be much less popular.

 

[pc1966]

This is an example of a decent voltage tester available in the USA:

https://www.homedepot.com/p/Klein-Tools-AC-DC-Voltage-Tester-ET45/308822920

NOTE: If you are going to work on something and intend on proving it is dead, you must verify the tester before and after you check out the circuit(s) you intend on working with. That can be done with a known live outlet, but better still you can get battery powered "proving unit" that generate a set of test voltages to verify the tester is working OK.

It is only about $9 and far safer and more certain than the non-contact type.