Arc Fault Breakers in a panel 3 wire feeder

Status
Not open for further replies.
I have a panel feeding a new bathroom addition. The home run is a 12/2 romex. feeding into a panel with a 3 wire feeder. So all the grounds and neutrals are landed on the same neutral bar. The installed breaker is functioning fine however my inspector is failing the job saying that the breaker will not function correctly due to the panel not having a dedicated ground in the feeder. But in my opinion this should not be an issue the breaker has no idea if the ground is separated or not. It is simply monitoring the return on the neutral and since this is a preexisting condition in a house over 40 years old it should not be cause for an inspection failure the new bathroom is wired to current code including the afci breaker. The home owner should not have to have their entire service and subpanel upgraded over 1 20 amp circuit. Looking for advice and if my line of thought makes sense.
 
If that's what the inspector is going on, he is wrong. As you said, an EGC in the feeder to the panel would have no bearing on the function of an AFCI breaker. Now some (most) mfg AFCI breakers have a form of GFI in them, I think along the lines of 30mA. But that would only be affected by the load side should there be a ground fault.
Also, if the house is existing and the 3-wire was legal when installed, it is "grandfathered" and doesn't have to be brought to current code.
I think you should ask to talk to the inspector's supervisor/boss.
 
. . . my inspector is failing the job saying that the breaker will not function correctly due to the panel not having a dedicated ground in the feeder. . .
If your inspector was right (which s/he isn't) an AFCI breaker in a service center panel supplied by Service Conductors (three-wire 240 / 120 Volt) wouldn't function correctly. But it does.

An AFCI breaker has no connection point to an Equipment Grounding Electrode, only the neutral and the hot.
 
A while ago I had one inspector question is the afci pigtail should be uncurled or not....:roll:, this repetition of outlandish conjecture hails directly from our trade's misunderstanding of afci functionality.

In my humble opinion, we need to keep this simple

It's chief operational device is a toroidal coil >>>

images


It can do no more/less than what a toroid can do

~RJ~
 
however my inspector is failing the job saying that the breaker will not function correctly due to the panel not having a dedicated ground in the feeder.

First question I would ask you, is the Neutral / Equipment ground Bus Bonded to the metal enclosure.

If not than i would clarify what breaker the inspector is referring to.

The inspector is not going to tell you to bond a bus to the panel enclosure that the NEC requires to be un-bonded.

If it is unbonded in my opinion the inspector is correct if you have a line to panel enclosure ground fault without a ground reference how are you providing ground fault protection

The inspector can tell you the NEC requires an equipment ground to provide that reference back to the service neutral
 
Last edited:
bonding

bonding

At the service the ground and neutral are bonded as per code it is only the feeder from there to the sub panel that does not have a dedicated ground only 2 hots and a neutral. At the sub panel the neutral/ground bar is not bonded to the can however all branch circuits are 12/2 with a ground so all grounds and neutrals are together. this is very typical for all the house in the area of this age built in the early to mid 80s or earlier.
 
afci

afci

Again in my understanding of the functioning of the breaker it has a neutral pigtail not a ground pigtail just like a gfci breaker. The breaker does not know or have a way of knowing if there is a ground or not in the event of a ground fault it will trip just like a standard breaker. As far as I know a gfci outlet is the code required solution for 2 prong outlet replacement in a 2 wire non grounded branch circuit. So it functions without a ground as well again just monitoring the neutral return.
 
At the service the ground and neutral are bonded as per code it is only the feeder from there to the sub panel that does not have a dedicated ground only 2 hots and a neutral. At the sub panel the neutral/ground bar is not bonded to the can however all branch circuits are 12/2 with a ground so all grounds and neutrals are together. this is very typical for all the house in the area of this age built in the early to mid 80s or earlier.

The subpanel must have the EGC and neutral bonded to the can since only a 3-wire was ran. And I bet that's what the inspector was trying to say. Nothing to do with the AFCI, it has to do with a fault to the can. There is no fault path back to the source should there be a fault to the can. Your main service is bonded this way, so why do you think the subpanel shouldn't be. Even if a 4-wire was ran, which is required today, the EGC is bonded to the can in a subpanel. Since you have no feeder EGC in the subpanel, the neutral must be bonded to the can, same as the service.
 
If the existing installation is only 40 years old, it wasn't right from the beginning,(needing a separate EGC with the feeder) unless the feeder in question is supplying a separate building/structure.

That said, the functioning of the AFCI is not affected by this.

If only thing supplied is in a bathroom, AFCI is not required, but sounds like it will be in 2020.
 
At the service the ground and neutral are bonded as per code it is only the feeder from there to the sub panel that does not have a dedicated ground only 2 hots and a neutral. At the sub panel the neutral/ground bar is not bonded to the can however all branch circuits are 12/2 with a ground so all grounds and neutrals are together. This is very typical for all the house in the area of this age built in the early to mid 80s or earlier.
Ddservices, you are describing a regional difference that is real interesting to me. I'm changing the topic a bit. . . and note that the AFCI breaker behavior answers we've given, above, are still the same. That is, an AFCI breaker won't care about anything more than the energized 120 Volt conductor and the grounded neutral conductor.

Do you, by chance, have a picture of the inside of a typical subpanel from that era wired as you describe in your posts?

My question goes to whether the subpanel can is actually unbonded, that is, actually floating, without any connection to a fault current return path. I find this fascinating. There is some history here that I'd like to know better.

Generally, in my personal experience, 1950s, and earlier, FUSE panels had their neutral terminal bars mechanically, and electrically, bonded to the enclosure by the manufacturer. However, based upon my memory and experience with the materials, the generic post-1950s CIRCUIT BREAKER panel has a manufacturer-supplied means to bond the neutral terminal bar to the enclosure at the installer's choice, and the neutral terminal bar comes from the manufacturer floating, electrically. So you are describing a local practice of not bonding the subpanel enclosure for possibly a period of 20-25 years . . . right? Is there any information about your area's local practice of this?
 
At the service the ground and neutral are bonded as per code it is only the feeder from there to the sub panel that does not have a dedicated ground only 2 hots and a neutral. At the sub panel the neutral/ground bar is not bonded to the can however all branch circuits are 12/2 with a ground so all grounds and neutrals are together. this is very typical for all the house in the area of this age built in the early to mid 80s or earlier.

this is very typical for all the house in the area of this age built in the early to mid 80s or earlier.

It is also typical through that era to be the same electrical inspector in an area, unfortunately as other said its not code compliant and the inspector either did not catch the violation or did not know any better.

Before you bond that bus at the sub-panel you need to talk to the current inspector he is correct that the sub feeder needed a forth wire equipment ground. The inspector may not approve that solution since he said you need an equipment ground.

If he requires the equipment ground the subject of separating the neutrals and equipment grounds may also need addressed.
 
It is also typical through that era to be the same electrical inspector in an area, unfortunately as other said its not code compliant and the inspector either did not catch the violation or did not know any better.

Before you bond that bus at the sub-panel you need to talk to the current inspector he is correct that the sub feeder needed a forth wire equipment ground. The inspector may not approve that solution since he said you need an equipment ground.

If he requires the equipment ground the subject of separating the neutrals and equipment grounds may also need addressed.
Correct that no EGC was and still is a code violation. Inspector was dead wrong to state the AFCI would not work though. It does not depend on any ground reference to function and would still function if you have an ungrounded system.

Nothing about running a new afci circuit should depend on the source here ~RJ~

Well if not a source that has 120 volts nominal AC we don't even need to consider AFCI's so far.;)
 
however my inspector is failing the job saying that the breaker will not function correctly due to the panel not having a dedicated ground in the feeder. .;)

Correct that no EGC was and still is a code violation. Inspector was dead wrong to state the AFCI would not work though. It does not depend on any ground reference to function and would still function if you have an ungrounded system.

From the information we have the inspector said the breaker will not function correctly, we do not know he said the branch circuit will not have AFCI protection

If the romex connector shorted out the ungrounded circuit conductor to the un-bonded panel enclosure no current would flow do to the lack of a reference to complete the short circuit

How is the inspector wrong to state the breaker won’t do its job without a equipment ground reference to the service neutral

We seem to be reading into this that the inspector said the arc fault protection won’t work, the branch circuit equipment ground is tied to the neutral bus there is an effective ground reference for the load side of the breaker past the panel enclosure

Edit this is not an ungrounded system though i think i get your point, but an ungrounded system has a bonded conductor to ground reference, i would think current would have to flow for a breaker to do its job
 
Last edited:
From the information we have the inspector said the breaker will not function correctly, we do not know he said the branch circuit will not have AFCI protection

If the romex connector shorted out the ungrounded circuit conductor to the un-bonded panel enclosure no current would flow do to the lack of a reference to complete the short circuit

How is the inspector wrong to state the breaker won’t do its job without a equipment ground reference to the service neutral

We seem to be reading into this that the inspector said the arc fault protection won’t work, the branch circuit equipment ground is tied to the neutral bus there is an effective ground reference for the load side of the breaker past the panel enclosure

Edit this is not an ungrounded system though i think i get your point, but an ungrounded system has a bonded conductor to ground reference, i would think current would have to flow for a breaker to do its job

As long as an MBJ exists somewhere, any fault can make back to the serving xfomer.

~RJ~
 
From the information we have the inspector said the breaker will not function correctly, we do not know he said the branch circuit will not have AFCI protection

If the romex connector shorted out the ungrounded circuit conductor to the un-bonded panel enclosure no current would flow do to the lack of a reference to complete the short circuit

How is the inspector wrong to state the breaker won’t do its job without a equipment ground reference to the service neutral

We seem to be reading into this that the inspector said the arc fault protection won’t work, the branch circuit equipment ground is tied to the neutral bus there is an effective ground reference for the load side of the breaker past the panel enclosure

Edit this is not an ungrounded system though i think i get your point, but an ungrounded system has a bonded conductor to ground reference, i would think current would have to flow for a breaker to do its job
Absolutely.

The AFCI does not need a ground reference to function. A ground reference may help with quicker development of conditions for it to respond to though.

This is no different then connecting that same breaker to an older wiring method that has no EGC, it will still do it's thing if the right conditions do develop.
 
If the existing installation is only 40 years old, it wasn't right from the beginning,(needing a separate EGC with the feeder) unless the feeder in question is supplying a separate building/structure.

That said, the functioning of the AFCI is not affected by this.

If only thing supplied is in a bathroom, AFCI is not required, but sounds like it will be in 2020.

I believe the 2014 NEC does not require AFCI in bathrooms. The 2017 NEC requires AFCI in bathrooms. (no need to wait for 2020) Do not know what NEC is being used by AHJ. Also, I agree that inspector can require OP to run a grounding wire to the sub panel.
 
I believe the 2014 NEC does not require AFCI in bathrooms. The 2017 NEC requires AFCI in bathrooms. (no need to wait for 2020) Do not know what NEC is being used by AHJ. Also, I agree that inspector can require OP to run a grounding wire to the sub panel.
Bathroom not in the list of rooms in 2017 210.12(A).
 
How is the breaker going to do its job, how is the breaker going to trip

Methinks there's some confusion, the OP panel is an older 3 wire w/bond to it's shell

Most are.

I'm willing to bet the meter , if similar vintage , is also , the GEC could exist at either, depending on the poco

~RJ~
 
Status
Not open for further replies.
Top