GFCI with no ground

[Scott Falvey]

Yesterdays inspection was on a 200+ year old home. House has combo of both 2 and 3 wire electric. I found 2 outlets that were GFCI, both did not have grounds (upgrades from 2-prong which by "code" is acceptable from my understanding).

They tested properly per the test and reset buttons which would indicate they are acceptable per the manufacturer, but when using my tester, they would not trip.

Is it acceptable if it tests properly using the device test buttons? and if so, how can it be acceptable if it will not trip with an outside source?

Thanks,

[David P. Valley]

Originally Posted by sfalvey

Yesterdays inspection was on a 200+ year old home. House has combo of both 2 and 3 wire electric. I found 2 outlets that were GFCI, both did not have grounds (upgrades from 2-prong which by "code" is acceptable from my understanding).

They tested properly per the test and reset buttons which would indicate they are acceptable per the manufacturer, but when using my tester, they would not trip.

Is it acceptable if it tests properly using the device test buttons? and if so, how can it be acceptable if it will not trip with an outside source?

Thanks,

Scott,

An ungrounded GFCI "Is Acceptable". A GFCI does not need a ground wire in order to trip properly. A GFCI simply trips when a fault is detected on the neutral wire. An ungrounded outlet will not trip using your tester. In this situation, I simply apply pressure to the GFCI test button in order to check its operation. And this outlet should be labled "GFCI protected / NO Equipment ground".

[Larry D. Kage]

Your tester uses the grounding conductor to test and since there is no grounding conductor your 3 light tester does not work. But, like David says, the integrated test button will test its functionality.

[Marc D. Shunk]

Seems like I recall some home inspectors making a "special tester" for this purpose, with the third prong on their GFCI tester going to a long wire with an alligator clip they clip on the nearest radiator or water spiggot. That rig probably is helpful for testing the ungrounded load side receptacles. I'm not that clear, myself, on how safe it might or might not be. Obviously, the UL hasn't given their blessing to such a home brew gizmo.

[Robert J. Wind]

If no ground; then the outlet must be labled saying that the GFCI is not grounded. They sell the stickers at any of the home improvement stores.

[Greg Fretwell]

I think the stickers are a joke. When my wife was building houses they had to go around putting "GFCI protected" stickers on to get the 305 and the first thing the owner did when they moved in was peel the stickers off.
The only one I have that actually survived was behind the door in the bathroom. All the ones in the kitchen came off within a month of normal cleaning.

[Robert J. Wind]

If you are inspecting a older home before GFCI and a GFCI outlet was installed then it should have the sticker. If the sticker is on the outlet then no issue. If you were conducting a code inspection (and i know that HI's do not do this) then it is requiered to have the sticker

[Ralph Brady]

I have a comment that states something to the effect of:
"The receptacle is not grounded and it has GFCI protection. Ungrounded GFCI receptacles should be labeled indicating lack of grounding."

On all open ground 3-prong and 2-prong receptacles, make sure you test for reverse polarity with an indicator probe.

[Eric Hyde]

I'm still confused on this issue as well. I read a post in this forum from 2006 that said that an NFPA official claimed that in Oct of that year the new GFCI's would not work without a grounding conductor.

I don't have access to the NEC or the NFPA codes, can anyone affirm or deny this claim?

[Larry D. Kage]

Originally Posted by ehyde

I'm still confused on this issue as well. I read a post in this forum from 2006 that said that an NFPA official claimed that in Oct of that year the new GFCI's would not work without a grounding conductor.

I don't have access to the NEC or the NFPA codes, can anyone affirm or deny this claim?

http://www.mikeholt.com/mojonewsarch...1~20020102.htm

[Eric Hyde]

Thanks Larry,

I appreciate the reply, but the link you provided didn't answer the question. Do you have one that does?

[Mike Whitt]

Originally Posted by ehyde

I'm still confused on this issue as well. I read a post in this forum from 2006 that said that an NFPA official claimed that in Oct of that year the new GFCI's would not work without a grounding conductor.

I don't have access to the NEC or the NFPA codes, can anyone affirm or deny this claim?

The NEC nor the NFPA would publish any thing like this it would be a NRTL such as UL.

The statement that a GFCI will not work without an equipment grounding conductor is just not true.
The EGC plays no role in how a GFCI works.

[Brian A. MacNeish]

Originally Posted by dvalley

Scott,

An ungrounded GFCI "Is Acceptable". A GFCI does not need a ground wire in order to trip properly. A GFCI simply trips when a fault is detected on the neutral wire. An ungrounded outlet will not trip using your tester. In this situation, I simply apply pressure to the GFCI test button in order to check its operation. And this outlet should be labled "GFCI protected / NO Equipment ground".

From "How Stuff Works":

A GFCI monitors the amount of current flowing from hot to neutral. If there is any imbalance, it trips the circuit. It is able to sense a mismatch as small as 4 or 5 milliamps, and it can react as quickly as one-thirtieth of a second.

http://ecmweb.com/mag/electric_gfcis_work/

[Larry D. Kage]

Originally Posted by ehyde

Thanks Larry,

I appreciate the reply, but the link you provided didn't answer the question. Do you have one that does?

I was thinking it would be helpful regarding how gfci's do work.