When I find a GFI outlet that does not trip with my tester, I note it as defective. Sometimes I find them that will trip from the test button, but not from the tester.
Anyway, last week I noted one as not working, and ungrounded, in the garage of a recently flipped home.
The electricians response is as follows “the reason the GFI in garage does not trip is because, the existing wiring in the garage is ungrounded. This is the same situation as ungrounded 2 prong receptacles in older homes being changed to GFI’s to get the three-prong configuration. This is why the electrical inspector will allow these GFI’s though a GFI tester will not trip them.”
First of all, is that correct?
Secondly, and only if it is, how is a person protected? If my tester does not trip them, how will a fault?
The electrician is correct. There are probably a hundred threads on this board discussing this matter.
Your “tester” requires a grounded (i.e bonded egc), three-wire circuit to test the internal function of a GFCI device. Not all existing dwellings have a three-wire branch circuit wiring system. Pre 1960’s homes (for instance) will rarely have an equipment grounding conductor as part of their branch wiring.
The GFCI device is not dependent on the egc for proper function. It’s circuitry detects an imbalance between the supply and return current of the alternating-current on the grounded and ungrounded conductors. Any minor imbalance should open the device.
Your “tester” creates this imbalance through the grounding conductor. If this conductor is not present, your “tester” cannot open the circuit.
On an ungrounded circuit, insert your “tester” into the receptacle, and push the internal test button on the GFCI device. If the power is turned off as a result, the device is functioning properly.
If I am testing a non-GFI outlet on an outside wall, and it does not trip, how do I know if it is protected (by a non grounded outlet elsewhere as you describe) or not?
Secondly, why do you instruct to insert my tester and then push the GFI test button, and not just push the button without my tester inserted?
If it’s a tree-pronged (grounding type) receptacle and is ungrounded, it’s required to be labeled “no equipment ground GFCI protected.” Otherwise, it’s a “defect.”
If you can’t verify the presence of GFCI protection, write it up.
You must always verify that there is no power to the receptacle. An improperly wired or defective device may allow for the internal button to trip, but not actually open the circuit.
Jeff is correct, but the electrician is suppose to label the outlet ungrounded. The outlet should have come with a decal “no equipment ground”. If the decal is missing, then call it out.
I have never seen the “Outlet does not have a ground” label. I have purchased numerous GFCI’s over the years and the only label in the box is the " GFCI protected outlet" Where do you get the no ground label?
boy do I ever not get it! and I definitely am no electrician per se but knew enough to rewire my entire house and not die in the process (though I experienced some enlightened moments) but theres my limited credentials. anyway my understanding is what mr pope stated that the GFCI measures the balance between supply and return current and if it senses an imbalance trips the unit, so what does the GFCI care where the current is going? (ie apparently requires a ground to be able to trip with the tester)…all it cares is that the out does not closely enough match the in. of course I believe the more knowledgeable than I (everyone here) are correct…I just assimilate concepts most effectively if I can understand the ‘how come?’ of the matter.
…and appear a bit more knowledgeable for I too have reported ‘non functional’ GFCIs on ungrounded systems.
mike in MN
hey roy, the illustration you provided is how I understood GFCI. that is…something, in this case a human body…becomes the earth ground and creates the imbalance which causes the GFCI to trip. that is what I thought I understood. so why does this only work in the case of a grounded system? I do not see an equipment or earth ground in the illustration
mike in MN
I was so elated thinking I understood and now I realize I only understand why the tester works. so I don’t get it why (JPs test) inserting the tester in the unit and trying to trip it would prove ‘functional’…after all the unit will trip with the ‘test’ button anyway. so in my limited understanding, if an electrician ‘fixes’ an ungrounded system by installing GFCI protection…you just accept that by faith? and theres no way to check it?
mike in MN
Your diagram does not show an egc, so yes, this circuit would show an open ground with any testing equipment.
Think about (and then go look) the appliances plugged into your kitchen-countertop receptacles and those on your bathroom countertop. I’ll bet you that most of them (if not all) are the two-pronged type, without an equipment grounding conductor.
If the GFCI device required an egc connection in order to be functional, it would be prohibited to manufacture small appliances without a factory installed egc.
The “testers” that we use for inspection purposes use a jumper from the ungrounded (hot) conductor to the equipment-grounding (egc/ground) conductor (through a resistor) to create an imbalance in the supply and return of the current. If the circuit has no egc, there is nowhere for the imbalance (“leak”) to go. Therefore, the tester cannot work on an ungrounded circuit.
The internal test button of the GFCI device creates an imbalance by redirecting the current through a 5mA resistor built into the device, and measures current running within the grounded (neutral) and ungrounded (hot) conductors, without relying on a grounding conductor.
How do we do this if the GFCI protecting the ungrounded circuit is located in the panel - protecting several ungrounded receptacles OR just the one we are trying to test?
Question
You mean that the tester will not work to test the GFCI but it will work for reverse polarity, hot/neutral reversed etc, correct?