GFCI & TESTER

[EI JOE]

Scenario:
GFCI Receptacle with one additional, standard receptacle DOWNSTREAM.

GFCI Tester (typical supply house $10 model) shows proper wiring via test lites in both receptacles.
Test Button on TESTER does NOT trip GFCI either from the GFCI receptacle or from the standard receptacle downstream.
The TEST BUTTON ON the GFCI RECEPTACLE trips the GFCI receptacle AND cuts power from the standard receptacle downstream.

Is my Tester bad (intermittant, as it does work on other GFCI's) or is there a wiring scenario here that is wrong but my tester cannot pick it up and I am being fooled?

 

[charlie]

Re: GFCI & TESTER

1.) Your GFCI tester may not work correctly (probable)

2.) There may not be a grounding conductor at the downstream receptacle (it will not work without one but the factory tester will operate properly without one)

3.) The grounding conductor has lost its continuity back to the source (see comment in 2)

4.) The grounded conductor has been used to loop to the grounding terminal to "steal" a ground (seems like it would work but the GFCI tester will not trip the upstream device)

I am not a fan of the GFCI testers but they have a definite purpose.

 

[charlie b]

Re: GFCI & TESTER

I can only understand the first item on that list as a possibility. My problem is that the GFCI tester does not cause a trip, when plugged into the GFCI receptacle itself. My understanding is that the tester creates a high (but not too high) resistance path between the hot and ground, causing a trickle current to flow. The GFCI receptacle should see that current, interpret it as a ground fault, and trip. That should happen no matter what is connected downstream. I drew a sketch, and I cannot see how any wiring error would both (1) Allow the lights to show a good connection on the downstream receptacle, and (2) Prevent the GFCI tester from initiating a trip.

If you have time to troubleshoot, I suggest disconnecting the load side wires (take the second receptacle out of the picture). I predict that the GFCI tester will still not work. My money is on the notion that you need to spend more money to get a new tester.

 

[al hildenbrand]

Re: GFCI & TESTER

It sounds like the trip current needed by the GFCI receptacle is a little high, and that the handheld tester isn't allowing quite that much to flow. And, for whatever reason, the builtin tester is meeting the trip current, allowing just a little more trip current than the handheld tester.

Do you have another GFCI tester that you can use in this test situation?

These little black box testers, whether built into the breaker or receptacle, or built as a standalone portable module, require us to take on faith that they are doing what we imagine them to be doing. Short of actually measuring the leakage current through the tester in the testing activity, there is no way to "know" what the tester is doing.

One possible wiring situation that could present the symptoms you describe is that of a high impedance ground. The handheld tester indicator lights are very high impedance, so that when the ground indicator is connected between the hot and EGC the resulting current doesn't trip the GFCI. If the actual impedance of the EGC is just high enough to not allow the 5 milliamp trip test current flow (by my quick calc greater than 24 K&#8486) it may still be low enough to allow the indicator light to work. Using a 3 prong plug adapter to a 2 wire receptacle (a "cheater") on the handheld tester would permit the running of a conductor from the cheater ground to a known ground somewhere in the building. If using the known ground allows the handheld GFCI tester to trip the GFCI, then the EGC quality is bad.

Another way to get at this would be to use an Ohmmeter to measure between the neutral and the EGC. It should only be as high as a few Ohms, ideally less than an Ohm. If the neutral to EGC resistance is low, the original handheld GFCI test results indicate the GFCI is bad.

 

[raider1]

Re: GFCI & TESTER

I ran into a problem similar to the one described here.

We had GFCI outlets installed in a shop, no slaves but a GFCI outlet at each location. I went to test everything before we had the inspection, and low and behold the GFCI test button would cause a trip but not my GFCI tester.

I went and borrowed an Ideal circuit analyzer/GFCI and AFCI tester. I analyzed the circuits and found a high impedance grounding path, all the lights on my simple tester showed good.

Long story short, I checked the transformer feeding the 208/120 panel and found that the transformer was not properly bonded. After I put in a main bonding jumper everything worked fine.

Chris

 

[al hildenbrand]

Re: GFCI & TESTER

One last thought.

One very important difference between the builtin test black box and the handheld GFCI tester is that the builtin test does not require a any EGC connection to operate and complete the test.

Edit typo - Al

[ October 06, 2005, 11:01 AM: Message edited by: al hildenbrand ]

 

[charlie]

Re: GFCI & TESTER

Charlie, it is my understanding that the plug-in style GFCI testers put a load between the ungrounded conductor and the grounding conductor. If that is the case, my scenarios are correct.

 

[EI JOE]

Re: GFCI & TESTER

It turns out that my hand-held tester is faulty. The ground pin is loose and the neutral is slightly loose. When I went to check other receptacles today, I started getting Open Ground & Open Hot readings all over a house and I could make the indicator lights change just by moving the tester around in the receptacle.

I have purchased a new tester.
A lesson learned...keep a [new] back-up tester in the bag as a 'check' against the first if something doesn't smell right.

Thanks to all for the quick replies.