No EGC- 3 wire feed

Circuit grounds that return to a panel (sub) that is not itself bonded to ground are indeed isolated. They are just a third wire that goes nowhere. The question that was raised by the OP was how was the sub panel bonded to grounds with no 4th wire feed and no visible connection between the panel housing and ground? The question arose because these circuits all tested properly at the outlets (which by itself indicates proper bonding).

The OP could have simply connected his solenoid tester from the ground bus in the sub panel to either hot leg to confirm that the “upstream connection” to the grounded conductor was intact.

The answer is, of course, as Robert Meier suggests.

3 Likes

As you’ve touched on there are many possibilities as to how this occurred. It is possible that the “extra” bare copper conductor in the sub-panel is being used as an EGC for the feeder. It is also possible that one of those circuits is connected to an appliance that is connected to a metal water pipe which is bonded. There are a few pieces of the puzzle missing like the possibility of the aforementioned junction box.

4 Likes

Bob, yes, but he already verified that they are not isolated. That’s not the question. A simple check with a continuity tester will verify whether they are isolated, which I presume he did. The conduit between the panel in question may have plastic conduit, but he said nothing about the other system components or how circuits were routed.

Electricians often indiscriminately tie grounds together, and unlike grounded conductors, there is no reason not to. In this instance, it may have even been done intentionally. I’ve seen a lot of strange ways of doing things.

The original question was how could they test as being electrically continuous. The answer is that they almost certainly were tied together with other grounds elsewhere in the system. I’ve seen installations where a ground was taken from a junction box to a water line or a gas line. Whether the grounds were tied in elsewhere intentionally or unintentionally, the fact that they tested as being electrically continuous is compelling evidence of them being tied in with other grounds that go back to an upstream panel.

I’ve disassembled many electrical systems, hundreds, while doing fire investigations or investigating and troubleshooting other electrical failures and accidents over a span of many years. I cannot recall a single time where I’ve encountered grounds that were truly isolated. The worst electricians seem to understand that they need to somehow carry the grounds back to the source. Whether they do it correctly or not is an entirely different matter.

2 Likes

When was the house built? If it was before the adoption of the 2008 NEC, there ought to have been a tie bar between the grounds’ terminal strip and the grounded conductor’s terminal strip. Three-wire feeds used to be quite common.

Why would you assume that? For the EGC connection?

That could be. But its likely that small, additional grounding conductor at the top right acting as the EGC.

If its not, then saying these grounding conductors are isolated in this subpanel would technically be correct. Because there is no bond between the grounded conductors and the enclosure, and not bond between this enclosure and the downstream enclosure. Only a bond between the grounding conductors and this enclosure, which is why I mention they dead end here.

Thanks robert. Assuming that is the EGC, would that be an issue? Too small, etc?

There are numerous NM cables entering both panels with the feeder. There are no visible branch circuit cables looped through the main panel so there has to be something between the panels besides the single raceway entering the smaller panel.

Possibly too small. I’m curious about how it gets there.

2 Likes

Is it possible there is not actually a continuous raceway, but just small stubs out of each panel?

1 Like

That’s possible. I believe that Daniel stated it was a raceway but short stubs would make sense.

1 Like

I am not sure… It is hidden behind drywall, but these two panels are side by side. I just commented that it isnt a metal conduit

1 Like

It’s certainly a mess the way it’s done. As Ryan suggested and I’m also guessing short stubs of PVC buried in the wall. So the real question (which comes back your OP) is the sub-panel properly grounded or not?

Yes. BUT #3 second breaker from bottom right does not have a wire connected. Dominic is correct.

1 Like

In your original post, you said that “Everything still tested grounded”. I took that to mean that you checked continuity between the grounds and the grounded conductors. If you didn’t, I don’t understand what you mean by “Tested grounded”. Please explain what you meant by that comment.

No offense but I suspect that you may not understand the concept of “Isolated”. In the electrical world, “Isolated” means not electrically continuous. Transformers, for example, all fall into two broad categories, isolation transformers and auto transformers. The transformers that serve buildings are isolation transformers. They are called isolation transformers because the primary is inductively coupled to the secondary. The two sets of windings are insulated and isolated from each other. If you were to check continuity between the primary and secondary with a standard DC tester, there would be no continuity. There are also devices called intrinsically safe devices that are used for isolation. They are used extensively in places like natural gas pump houses. Unfortunately, the term “Isolated” is one of the most misused electrical terms by home inspectors. Being insulated, separated, or segregated at a particular location does not make conductors isolated. They are isolated ONLY if there is no electrical continuity between them.

The first thing to do, therefore, would be to check for continuity. If there is none, then yes, they are isolated. However, if there is continuity, then the grounds from the sub-panel circuits are mechanically connected to a common ground elsewhere. Probably, that common ground is another EGC elsewhere in the system.

When electricians wire a house, they often run segments of circuits to junction boxes and then a home-run from a junction box back to the panel where the circuits originate. When they do, they tie all the EGCs together in the junction boxes. They should take care to keep grounded conductors together with their associated circuits, but in my experience, they don’t. They tie all the grounded conductors together. You can be sure that if they don’t properly segregate the grounded conductors, they certainly aren’t going to segregate the EGCs. As I said before, though, they do not need to segregate the EGCs. I’m merely describing what I see on a regular basis.

Sometimes in attics, crawlspaces, and basements, conduit is used instead of NM cable for portions of circuit runs. Say, for example, a large ranch style house has the service or a subpanel a long distance from many of the circuits. The electrician might run a 3/4" EMT for a significant portion of the distance with several junction boxes along the run. That way, the electrician saves both time and money by running fewer conductors than there would be if all NM cables were used. Instead of, say, 10 or 12 home runs in NM, each with its own EGC, the EMT could be used as the EGC in lieu of the multiple EGCs in cables. Each junction box is going to have all the EGCs tied together and a pigtail to the junction box itself. That virtually eliminates any possibility of there being any isolated EGCs anywhere in the house.

In the city and county where my electrical contracting business was based, we had local codes that required EMT, IMC, or RMT in locations that the NEC does not. I also worked a lot in Chicago and they had similar rules. I’ve worked in other jurisdictions where the NEC permits the use of NM cable but local codes do not. The conduits further decreased the possibility of the existence of any isolated grounds.

To be clear, even if the circuits are all run in NM cable, the probability of the EGCs from any panel being isolated are infinitesimally small. I could see where a do-it-yourselfer might install a small two or four circuit panel in a detached garage or shed where they didn’t have a clue about what they were doing and end up with isolated EGCs, but I’ve never encountered such an installation. The do-it-yourselfers almost always, close to 100% of the time, tie the grounded conductors and grounds together in small remote sub-panels. In fact, that was a common practice for electricians until relatively recently.

So, while it is technically possible to have the grounds isolated, it just never happens in the real world.

I just meant with my plug in tester, at all outlets.

Thanks for taking the time to respond all that. But yes, I do understand the terms bonded and isolated. I was just stating that if there was no EGC in this panel, and there were no upstream connections of grounded and grounding, then it would have been isolated. (which is what it appeared to be at first, without knowing if there are any upstream bonds somewhere)

But thanks for helping me think through all the possibilities! Next time I will be sure to use a continuity tester. I am just not in the habit of sticking things into the panels… :grinning:
Not that I am afraid or unqualified, just dont usually need to for a general home inspection…

2 Likes

And you’re correct, if the feeder did not have an EGC there would be no connection between the EGC bus in the panel or the downstream receptacles that you tested and the egc bus in the panel where the feeder originates. Since there appears to be 12 cables entering the sub-panel are you making the reasonable assumption that there is a proper EGC run with the feeder and that the panel is properly grounded? How would you write this up in your report?

Yes, that is what I assumed. So, I actually chose not to mention it in the report. Since everything tested grounded, (with the plug tester) and it appears to be an EGC since it is larger than all the other circuits, I didnt want to recommend they pay for an electrician if not necessary.
The only thing I wasnt sure about calling out or not was the size of that EGC…

4 Likes

There are some “continuity” testers that can’t be used on live circuits and you can’t use the “continuity” setting on your DMM on a live circuit.

I would just use a DMM on voltage setting, or a wiggie…just sayin’ :wink:

A continuity tester will not work in this scenario as it is simply impractical.

To verify a grounds integrity either at a panel or an outlet you simply need to place your wiggy from the ungrounded conductor (hot) to the ground conductor. If the ground is bonded to the neutral at the service panel as it should be you will get a reading. If, on the other hand, there is no reading you don’t have a neutral/ground bond (one or the other is open - typically the ground). If there is no reading you should verify your “hot” connection by measuring from hot to neutral.

2 Likes

Exactly! That’s why I wrote “continuity” in quotes.

“Test for continuity” would have been a better statement, IMO, instead.