GFCI and an open ground

[fqp25]

Can GFCI's work properly with an open ground. Did an inspection this morning and my little outlet polarity tester gave me a reading for open grounds for most of the outlets. This tester also has a little "GFI Popper", but I can't remember if it worked or not due to the open ground.

[hausdok]

Hi,

Yes, if a ground fault occurs the GFCI will trip - even without a ground. It's measuring the amount of current carried into and out of the device. If it senses a difference between incoming and outgoing voltage that is beyond the tolerance it is set for, it will trip.

Lots of pre-1962 homes have the original ungrounded 2-hole receptacles converted to the 3-hole type receptacle without having added a ground. Placing a GFCI receptacle in the first outlet in the voltage stream, converting all outlets downstream to the 3-pronged type and then labeling them "GFCI Protected" isn't the ideal, but it's better than nothing and a reasonable solution that can give a repair-cash-strapped buyer some wiggle room, by making things safer until such time as he/she can afford to have all of the receptacles upgraded with equipment-grounding conductors.

ONE TEAM - ONE FIGHT!!!

Mike

[Tim Moreira]

Can GFCI's work properly with an open ground. Did an inspection this morning and my little outlet polarity tester gave me a reading for open grounds for most of the outlets. This tester also has a little "GFI Popper", but I can't remember if it worked or not due to the open ground.

With all due respect as I am new to this forum, I do not believe that the $10.00 tester from Lowers or HD is the *best* electrical tester to use for this type of application. I think that as HIs we should us something along the lines of a Suretest or other similar device. More expensive, yes, but more accurate. IMHO.

[hausdok]

Well,

With all due respect right back atchya, the SureTest will not trip a GFCI or work beyond checking voltage, polarity and voltage drop without a ground. Without a ground, it's not a whole lot different than the $12. testers that I abhore. I know that because I've been using a SureTest since 1999.

OT - OF!!!

[Martin Morgan]

When I come across this I simply say that the GFCI is not installed according to the manufacturers installation guidelines which supersedes all codes

see attached

[Jim Katen]

Originally posted by Martin Morgan

When I come across this I simply say that the GFCI is not installed according to the manufacturers installation guidelines which supersedes all codes

see attached

I just read through the installation instructions for a GFCI receptacle and I can't find where the instructions prohibit this. What was it you were going to attach and what does it say?

- Jim Katen, Oregon

[Martin Morgan]

I cannot get the pdf file to upload. send me an email.

allbay@yahoo.com

I will return it with the PDF file attached

[hausdok]

Hi Martin,

It's probably too large. Since it's a pdf, you should be able to reduce it. Have you tried using an image resizer on it to get it under the upload threshold? Also make sure that there are no special symbols or spaces in the file name. For instance a file name like GFCIinstructions will work fine but GFCI(Instructions) or GFCI Instructions won't upload.

OT - OF!!!

M.

[pi]

Ungrounded GFCI outlets are fine and common in old ungrounded systems, but MUST be labeled "No Equipment Ground". If a homeowner does not know it is ungrounded and plugs in a surge protector, their $10,000 worth of electronics could go poof from a line surge because surge protectors need a ground to work. Problem is, most homeowners don't know this either so we must kindly educate them. Also, when when checking these, tester will not trip them, but they still must trip when you press the GFCI "test" button.

[aelliott]

Going back to the old two prong type outlets; Is it accurate to say that three terminal type appliances cannot be properly pluged in to two prong style outlets. Some home owners will buy an adapter to plug these type appliances in. They have a false security that the appliance is properly grounded.

[hausdok]

Hi,

That might work in Kurt's neck of the woods where everything is in conduit and the center screw provides a ground, but it's pretty useless anywhere else.

Tell 'em to either upgrade, add GFCI's or make sure they only use double-insulated devices (the ones with a two-pronged plug) in those receptacles.

OT - OF!!!

M.

[Jesse]

Yeah, the adapter would work in a two-prong outlet in the Chi-town area. That is, of course, assuming the run to the receptacle was done properly AND the user actually utilizes the little pigtail on the adapter to the center screw. Never seen that done.

Around here, test the receptacle for ground. If it's good, let the client know the same box can handle today's three-prong receptacles.

[Brandon Chew]

Originally posted by aelliott

Going back to the old two prong type outlets; Is it accurate to say that three terminal type appliances cannot be properly pluged in to two prong style outlets. Some home owners will buy an adapter to plug these type appliances in. They have a false security that the appliance is properly grounded.

Here is my boilerplate on this issue. Feel free to steal it or to help me tweak it to make it better.

There are many two-slot receptacles present in this home, which is common for homes of this age. Appliances and equipment which come from the factory with two-prong plugs can be safely used in these receptacles provided the receptacle is wired with the proper polarity. However, many modern appliances and equipment such as refrigerators, microwave ovens, computers and other electronic equipment require an equipment ground for proper and safe operation and they are provided with three-prong plugs by the manufacturer for this purpose.

There are commonly available three-prong to two-prong plug-in receptacle adapters (called ground lift adapters) that are sold to consumers, however these still require a connection to a good ground near the receptacle in order to properly protect people and the equipment being plugged into them. In practice, proper connection to a good ground is rarely done by people using these adapters and I do not recommend their use. If a good ground was present near the receptacle that could be easily connected to the adapter as required, then it would be a relatively simple matter for an electrician to permanently install a three-prong receptacle at the outlet, negating the need to use the adapter.

IMPROVEMENT RECOMMENDED: Check areas of the home where you may wish to use equipment having three-prong plugs and have a qualified electrician properly install three-slot receptacles where needed. Be aware that installing these receptacles may require the addition of new branch circuit wiring with grounding conductors if a proper means to establish an equipment ground is not present at the existing outlet boxes."

I don't like the "replace with GFCI and label 'No Ground'" option so I don't mention it. Even though it is labeled "No Ground", if someone has a three prong plug, and they want to plug it in, and there is a three slot receptacle nearby, then they are going to use it no matter what the label says. The GFCI will protect people from getting shocked but will not protect the equipment that needs a ground. The electrician can discuss that option with the client if he wishes to, while he is "properly installing three-slot receptacles where needed".

[housewhisperer]

A non-GFCI overcurrent protection device (circuit breaker or fuse) is designed and intended to clear the fault by disconnecting only the hot conductor from the circuit when an overcurrent condition occurs. However, GFCI devices (both receptacles and circuit breakers) are designed and intended to disconnect both the hot conductor (also referred to as the “ungroundedâ€

[Brandon Chew]

Originally posted by housewhisperer

Lastly, consider the following scenario. You are inspecting an older home which is still wired with two-conductor wiring (no equipment grounding conductor). To provide a measure of safety against a potential ground fault condition, the original two-wire receptacle on the exterior of the home has been replaced with a GFCI receptacle and a metal weatherproof receptacle cover has been installed over the receptacle.

In this type of installation the metal screw which secures the metal weatherproof cover plate to the receptacle is screwed into the threaded hole in the metal portion of the receptacle. In turn, the metal portion of the receptacle is electrically continuous with the equipment grounding pin contacts in the grounding pin holes of the receptacle. So the cover, the screw, and the grounding pin contacts in the receptacle are all electrically continuous. Remember, this is wired into a two-conductor circuit so the grounding portion of the receptacle is not connected to any grounding source.

Now, being a thorough inspector, you kneel on the patio or ground in front of the receptacle and hold the little spring-loaded cover/door over the receptacle open with one hand at the same time insert your push button GFCI receptacle tester into the receptacle with your other hand and push the “testâ€

[housewhisperer]

Brandon,

Thanks for your question. Your confusion is understandable and useful. If someone doesn't understand what I say, it's my responsibility to look at what I write or say and to clarify my explanation.

Remember, in the situation I described, a GFCI receptacle is installed in an older two-wire system where no equipment grounding conductor is present. I didn't say that the GFCI device wouldn't trip when current passed through the inspector's knee where it contacted the concrete (we know that current doesn't just take the path of least resistance to ground, it takes all available paths to ground). I said that the inspector would get a shock (slight though it might be).

Now, consider this same scenario but with a failed GFCI device. Unlike the new requirements for GFCI devices which specify that such devices not allow any current to pass if their internal transformers fail, the ground-fault protection function of older GFCI devices can and does fail while still allowing the receptacle to remain live. If this were the case in the situation I described, the inspector would, at the minimum, experience a 120V shock and, at the worst, be electrocuted.

The point was to remind inspectors that understanding the function and limitations of their testing equipment as well as the potential hazards with respect to older GFCI devices installed in two-wire systems is important.

Thanks, again, for your input. I hope this helps. All of us benefit when we ask questions and share our knowledge and experience.

[Joshua]

HouseWhisperer (great name BTW)

I'm agreeing with most of what you've said. I am wondering about this one:

However, GFCI devices (both receptacles and circuit breakers) are designed and intended to disconnect both the hot conductor (also referred to as the “ungroundedâ€

[Brandon Chew]

Thanks for the response. Here is what was confusing to me. You began by saying "Now, we will discuss situations where a fully functional GFCI device will not provide protection from shock or electrocution." Then there were three examples. The first was the human filament. The second was hitting a secondary power source. The third, assuming the GFCI was functioning properly, is an example of the GFCI doing what it is designed to do. In that case, you wouldn't get a shock any worse than you would get from a normally functioning GFCI, would you? I agree if the GFCI malfunctions then there is a danger of electrocution. There is also a danger of electrocution in that scenario if instead of a GFCI receptacle, joe homeowner wired that exterior outlet with a three-prong receptacle on a two wire circuit (without equipment ground), and there is no GFCI breaker. Now, whether or not you see a "no ground" label on the receptacle, if you are careless and plug in your GFCI tester and press the button (thinking exterior receptacles are supposed to be on GFCI), you run the danger of getting a nasty shock just like you described. Lesson here would be if you plug in your three light GFCI tester and it lights up "open ground", don't push the GFCI test button on your little tester. You are about to send some current out the ground pin on your tester that is going to be looking for a way back to ground.

[Jim Katen]

Originally posted by housewhisperer

A non-GFCI overcurrent protection device (circuit breaker or fuse) is designed and intended to clear the fault by disconnecting only the hot conductor from the circuit when an overcurrent condition occurs. However, GFCI devices (both receptacles and circuit breakers) are designed and intended to disconnect both the hot conductor (also referred to as the “ungroundedâ€

[housewhisperer]

Jim,

Thanks for catching my typo error. You're correct when you indicate that it's "or" not "and." I'll proof my posts with greater care in the future.

With regard to the "fully functional" GFCI receptacle installed in a two-wire system - as I tried to clarify for Brandon, in the situation described with the inspector creating the path to ground for fault current, in this secnario the GFCI will typically trip when the cube tester button is pushed precisely because, when it tries to induce a hot-to-equipment grounding conductor fault on the non-existent equipment grounding conductor, the inspector will serve as the path to ground through his contact with the metal exterior receptacle cover and his knee on the concrete. He will experience a very slight shock. However, if the GFCI device is a n older type and its internal GFCI function has failed, the receptacle will still be live and there will be no GFCI protection for the inspector. He may as well kneel on the concrete and directly touch the hot conductor.

I'm pleased to see that this has generated a broad discussion.

As for any inspector disconnecting the white pigtail of a GFCI circuit breaker to test for grounded conductor (neutral) continuity, NEVER ON AN INSPECTION and at home only if you're thoroughly versed in electrical work.

[housewhisperer]

Just an addendum to my recent post. If any of you would like more information to help you understand bonding and grounding as well as the reason that, with few exceptions, equipment grounding conductors and grounded (neutral) conductors must be kept separated and the grounded (neutral) conductors "floated" in any enclosure other than the enclosure containing the main service disconnection device, contact me at prospex@prospex.us and I'll send you an illustrated article on these topics.

Kevin