GFCI test

[Oldmaster2]

Installed a dedicated 20 amp. GFCI circuit in kitchen. Used GFCI breaker in panel. Inspector tested it with plug in circuit tester and it did not trip. Breaker does trip with test button. I checked everything and I'm sure the wiring is correct. I noticed that instructions with breaker recommends to test with button on breaker. Is it possible that the plug in tester just won't work with GFCI breaker?

 

[tom baker]

Yes I believe you are correct.
Plug in tester is not the mfg approved way to test a GFCI, only use the test button. Normally with a receptacle a plug in tester won't work if there is no EGC. It appears a plug in tester won't work with a CB?

 

[Oldmaster2]

I'm positive there is a functional equipment grounding conductor. All the wiring is new.

 

[curt swartz]

You used a GFCI breaker or AFCI breaker?

 

[synchro]

As Tom mentioned, a plug-in tester needs an EGC because its test button simulates a ground fault by connecting a resistor between hot and EGC.

If you have solenoid tester (Wiggy) then connect it between the hot and EGC of a receptacle on the circuit. This should trip the GFCI.
If it doesn't trip, then measure the voltage between the neutral and EGC while the Wiggy or other small load is connected from the hot to EGC (to apply some current to the EGC). If there's a substantial N-EGC voltage then there is a problem with the EGC.

Is this a single pole GFCI circuit? That's likely, but I'm asking just in case it's a 2-pole GFCI breaker with a load neutral for a MWBC. With a MWBC it's possible that there could be some leakage on the other phase that's not enough to trip the GFCI by itself, but which could cancel enough of the leakage applied by the test button to prevent the GFCI from tripping. To do this the leakage on the other phase would have to be around 1mA to 5mA. This is a fairly narrow window and so it's not very likely to happen, but still possible.

 

[Oldmaster2]

Single pole

 

[don_resqcapt19]

I'm positive there is a functional equipment grounding conductor. All the wiring is new.

A lack of a solid EGC path, is about the only reason that a plug in GFCI tester won't open the upstream GFCI device...assuming that the plug in tester is functional.

 

[Oldmaster2]

Actually I was taught that a GFCI will work without an EGC . I know that's true with GFCI receptacles.

 

[Little Bill]

Actually I was taught that a GFCI will work without an EGC . I know that's true with GFCI receptacles.

They will work without an EGC, what Don said/meant was the plug-in testers won't work without an EGC. They shunt some current to ground terminal to trip the GFCI. Without an EGC, there is no path to ground.

 

[Little Bill]

I ran into a situation with GFCI receptacles that had a functioning EGC. A plug-in tester would not trip them but the test button would. Since you can't explain this to most Home Inspectors, I changed the GFCI. The new ones would trip from a tester. I ended up changing about 6 of these. I discovered that all of them were Harbor Freight brand. I didn't even know they made/sold GFCI. So I was convinced that they were just defective GFCIs.

 

[don_resqcapt19]

Actually I was taught that a GFCI will work without an EGC . I know that's true with GFCI receptacles.

Yes, GFCIs work just fine to provide shock duration limitation without an EGC.

They will trip where there is a path back to the system that can flow the 4 to 6 mA of current that is required for the GFCI to open the circuit. This path can just about anything that is conductive and in contact with earth or in contact with something that is connected to the electrical grounding system. Of course one such path would be an EGC, but there are a multitude of other paths for this current to flow on.

Where you use a plug in tester, the push button connects a resistor between the ungrounded conductor and the EGC at the receptacle. If there is no EGC, there is no circuit, and no current will flow, and the tester will not cause upstream GFCI to open the circuit.

 

[winnie]

Is it possible that the plug in tester just won't work with GFCI breaker?

As others have noted, a plug in tester requires a functional connection to ground in order to simulate a fault.

If you have a good path to ground, and if the plug in tester is functioning correctly, then the breaker should trip.

One possibility: if the breaker is functioning correctly, and the plug in tester is properly simulating leakage just above the trip threshold, then the breaker might not trip in the time the test button is pressed.

Many GFCIs trip immediately upon detecting a current above their trip level, but I believe the standard allows them to take several seconds at 6mA.

Jon

 

[don_resqcapt19]

...
Many GFCIs trip immediately upon detecting a current above their trip level, but I believe the standard allows them to take several seconds at 6mA.

Jon

That is correct. UL 943 would permit the trip time to be 5.59 seconds for a 6mA ground fault.
The standard gives the trip time as the quantity 20 divided by the ground fault current raised to the 1.43 power
(20/current)^1.43

Most trip much faster than that, but they would be compliant with the standard if they took that long.

 

[rnatalie]

Probably his tester is busted. By "inspector" do you mean a guy from the AHJ or some "home inspector?"

 

[Electromatic]

I can just imagine someone (cough..home inspector) cutting the ground pin off a receptacle tester to test 2-prong receptacles and trying to use that same tester to trip GFCIs.

 

[480sparky]

I've had a couple instances where a plug-in tester will not trip a new GFCI install. The built-in tester trips it, but not the plug-in.

 

[augie47]

I've had a couple instances where a plug-in tester will not trip a new GFCI install. The built-in tester trips it, but not the plug-in.

I'm depending on my memory which is a dangerous task at this stage but as I recall on some of the older GFCI receptacles you could press the "test" button and it would "trip" but not actually turn off (happened when they were wired with Line-Load backwards). I questioned Leviton and the answer was that was why the instructions (those things we all read) said test with a load plugged in.
The newer (relative term at may age) GFCIs don't have that problem.

 

[winnie]

I'm depending on my memory which is a dangerous task at this stage but as I recall on some of the older GFCI receptacles you could press the "test" button and it would "trip" but not actually turn off (happened when they were wired with Line-Load backwards). I questioned Leviton and the answer was that was why the instructions (those things we all read) said test with a load plugged in.
The newer (relative term at may age) GFCIs don't have that problem.

With the newer GFCIs, the 'test' button simply turns off the power, with no testing at all. The test happens when the reset button is pressed.

-Jon

 

[Dennis Alwon]

He said the gfci test button on the breaker did not trip the breaker. Bad breaker or if I recall you must hold the button on the breaker down for 3 seconds

 

[retirede]

He said the gfci test button on the breaker did not trip the breaker.

I don’t think that’s what he said.
Test button DID trip...plug-in tester did not.