Troubleshooting possibly bad GFCI circuit breaker or bad wiring

Question

I have a GFCI circuit breaker in our house (120 V, US) that tripped and would not reset, and I'm trying to investigate if the breaker is bad, or if there is a short somewhere in what it controls/connects to.

It is labeled as being for bathroom outlets, I have made sure nothing is plugged in to those outlets, and to investigate I disconnected the circuit breaker. After disconnecting, I could turn it back on, so I reconnected it with an ammeter in series and measured ~30 mA. It seems it should be zero unless one (or more) of the following is true:

• it controls some other outlets / devices etc. besides the known bathroom outlets
• this is within the "noise" of measurement or acceptable limit for house wiring
• there is effectively a ~4 kilo-ohm short in the system

I made some other measurements as sanity checks:

• this problem circuit but with vacuum cleaner plugged in and on: ~10 A
• main bedroom circuit: ~20 mA
  • only has alarm clock plugged in
• refrigerator circuit: 1.6 A
• washing machine: ~17 mA

Any ideas / suggestions for how I can determine if the circuit breaker is bad or if there is a problem with the wiring or one of the outlets?

Answer 1

Modern Code limits what that circuit could serve

If the installation is in the last 20 years or so, modern Code would apply, and that limits the possible uses of a bathroom receptacle circuit. It must pick one of these rules and follow it:

• the circuit may serve ONLY receptacles ONLY in bathrooms plural.
• The circuit may serve ONLY receptacles in 1 bathroom, but can serve hard-wired loads in that same bathroom also.

So that is what I would expect in a house modern enough to be sold with a GFCI breaker.

GFCIs care about difference in current

Current flows in loops. In a perfect world, hot current should exactly equal neutral current. GFCIs work very simply: they compare hot and neutral current, and determine the difference in current (which should be zero). >5mA and they trip.

You are measuring current on a hot wire or a neutral wire. One is meaningless without the other, at least as far as troubleshooting a GFCI. If you are measuring 30mA on the hot wire, and unbeknownst to you there is also 30mA on the neutral wire, that is working as intended. (Since the GFCI is holding, neutral current must be >25mA or <35mA, yes?)

What will trip a GFCI is >5mA current from hot to something else or neutral to something else. I would say "ground" but I don't want you to stick an ammeter on the ground wire. It may be transiting the water pipes to the system grounding electrode and back that way. Or current may be moving from this circuit's hot to another circuit's neutral, or vice versa, because a nitwit borrowed a neutral (a code violation even if GFCIs aren't involved).

What should test out

If you disconnect hot and neutral wires from the GFCI, (but leave the curly neutral pigtail on it), a couple of things should be true.

First, the GFCI breaker should hold. If the GFCI (any type) tripped with no "Load" wires on it, that's a defective GFCI. The GFCI holding is a very positive sign that it's not the $40 GFCI.

Second, if you gather hot-neutral together and connect them to each other, and measure that to Ground, there should be no conductance at all. That is to say, infinity ohms. You're lucky if a cheap DVM fails the circuit at this point, because that gives you something to test for. But many insulation failures behave like a VBO - insulate at <=X voltage but conduct above X voltage. To detect those, you need an insulation tester aka a "mega-ohm-meter" or "MEGGER" which is an ohmmeter with a very high test voltage of 250, 500 or 1000 volts depending on machine.

Fortunately, your panel already contains a MEGGER with a 120V test voltage and a 5mA detection threshold! Yeah. That GFCI.

In fact, if you wanted to gather the circuit's hot and neutral and put them both on the hot terminal of the GFCI, that would be a fairly effective insulation test! (it would put a higher test voltage on neutral than it normally gets). That won't hurt any of the equipment.

The next step, then, is to figure out the circuit's route and break it (both hot and neutral) near the midpoint. Test with only half the circuit connected. Does it still trip -> it's in the first half. If it holds -> it's in the second half. Divide the circuit at the 1/4 or 3/4 point (depending on the results of the first test) and test again. Continue to "divide and conquer" until you have identified the device or cable run at fault.

Answer 2

I do not know your resources or skill set so I will take a SWAG: From what you are saying I think you have a fault in the electrical part of your system not the breaker. I am assuming nothing was changed in your system before this started to happen. If for any reason a ground or a neutral from another circuit contacts this neutral it will trip. This approach is a big brush, there are a lot of details that may show uIf that is not the case start by removing the wire from the breaker. Then turn the breaker on, if it trips it is the breaker, if not it is the wiring.

When these breakers first came out they were set for about 5mA and noise sensitive. If it is an old one it should trip because of the 4K ohm fault, most new onis will also trip at this level, it is about 30mA. To isolate this, start by unplugging and/or disconnecting everything on that circuit (hint it will be off as the breaker is tripped). Connect the wire back to the breaker. Turn on the breaker. If it trips it is in the wiring, if not it is something that is connected to it.

If it did not trip you can start by connecting each load one at a time until it trips, you then know where the problem is. If it doesn't then you will have to start removing receptacle etc and isolate sections of the wiring. At any point during this process you become uncomfortable ask for help ans/or hire an electrician. Be safe so you can come back and tell us you found the problem.