Bonding Older 3-wire sub panels

[Ed Battle]

I am inspecting the electrical system of an older home, of which the owner does not necessarily want to re-wire, but nevertheless I want to provide earthing and bonding to make the electrical system as safe as possible. This in an older Victorian-era home with many rooms. The distribution consists of two main panels, and a system of 5 sub panels throughout the premises.

I am designing a good grounding system in the main distribution section following the guidlines of the NEC. It is the sub panels that I have some confusion on where to go.

The sub panels throughout this home are fed with 3 wires (2 hot conductors, and a neutral). No third grounding conductor was pulled. If there is a fault imposed on the chasis of one of these panels (or any other apparatus supplied by these panels), there is no path to clear this fault back to the source. I realize that the simplest way to solve this dilemma is to run a ground conductor from these panels back to the established ground at the main distribution system, but this may not be practical due to the architecture of the house.

My question is: Can I bond the neutral to the case of the sub panels to establish a path to ground? I realize that this is not acceptable under normal circumstanses, but in this case I cannot find any other method of grounding the equipment in this facility. Are there exceptions to the code in this circumstance?

Ed Battle

 

[infinity]

My question is: Can I bond the neutral to the case of the sub panels to establish a path to ground? I realize that this is not acceptable under normal circumstanses, but in this case I cannot find any other method of grounding the equipment in this facility. Are there exceptions to the code in this circumstance?

Ed Battle

I know of no exception that would permit you to do this. What is the wiring method used to provide these feeders to the subpanels? If they're metallic conduit then they may qualify as the EGC. Otherwise I'm sorry to say but it sounds like a rewiring job is in order.

 

[al hildenbrand]

This in an older Victorian-era home with many rooms. The distribution consists of two main panels, and a system of 5 sub panels throughout the premises.

The sub panels throughout this home are fed with 3 wires (2 hot conductors, and a neutral). No third grounding conductor was pulled.

Are the sub panels fuse centers or circuit breaker centers?

If the subs are fuse centers, are you sure that the neutral bus isn't already bonded to the "chassis" by the original manufacturer?

Or are the sub panels site assembled porcelain fuse holders in an early 1900s Knob & Tube panel?

The answers to these, and related, questions helps to establish the "archeology" of the system by indicating the original installation Codes that probably applied. The history of bonding is not monolithic. Bonding has evolved over the 20th Century.

 

[LarryFine]

Only other possibility is to make the panels 120v instead of 120/240v and re-arrange the conductors to use one as the EGC.

 

[al hildenbrand]

Based on the information given in the opening post, the assumption that bonding the subpanel pans will provide an "electrical system as safe as possible" is questionable.

 

[haskindm]

NO, NO, NO! Neutrals and grounds MUST be kept separate in sub-panels. Otherwise neutral current is imposed on all of the grounding conductors and the metal panel enclosures. In the event of a bad neutral connection these items would be energized to system voltage with no way to clear the fault. This was probably in violation when it was originally installed, but I don't know that for sure. New feeders will need to be pulled to these panels to make this installation compliant. There is no other way.

 

[al hildenbrand]

There is no other way.

Based on what I'm reading in the opening post, this is an OLD system. The metal of the subpanels apparently is NOT bonded to the neutral, and is floating without connection to the GEC.

If there is a fault imposed on the chassis of one of these panels (or any other apparatus supplied by these panels), there is no path to clear this fault back to the source.

In my area, I routinely find early 1900s original K&T 30 Amp 240/120 Volt single phase plug fuse service centers that are housed in black painted steel boxes. These K&T panel boxes were never bonded to anything resembling a GEC and they were never bonded to the service grounded conductor.

Over the decades after the original installation, an additional circuit or three would be extended from the panel. In my area, the first most common was a separate circuit for the first furnace replacement, the replacement now requiring electricity to run. If the wiring method was metallic (most cases it's armored cable fished up to the service, and conduit across the basement) the "earthing" of the service center box occurs through the gas and/or water pipe connections to the furnace.

 

[benaround]

Next question, do the branch circuits leaving these "sub panels " have EGC ?

It would not help to replace the feeders if only 2 conductor circuits are used.

 

[haskindm]

I guess what I should have said is: If there is no grounding conductor in the branch circuits and the neutral is isolated from the case in the sub-panel, then leave it that way. Any attempt to "improve" the system by trying to use the neutral as an equipment ground would actually introduce additional hazards. Many of the early electrical systems did not include equipment grounding conductors. The problem is when we try to "upgrade" the system and mix grounded circuits with ungrounded panels. Either leave it alone or start over, there is no halfway patch.

 

[Dennis Alwon]

Ed--- Are we correct in assuming that the sub panels are fed with cable without an EGC or is it a metal conduit feeding these panels with no EGC.

If it is cable then those panels have no ground and cannot be changed unless another wire is installed. The neutrals cannot be bonded to the panel at the sub panels.

 

[augie47]

sub-panels

sub-panels

wow! sometimes I feel I'm in a different world. In these parts, it is not uncommon to change out the main panel, reconnect the sub panels, make a "safety check" (open splices, etc) and be gone. Inspection approved.
Here we check "what was installed"....if the existing wiring to the sub panels is left intact and only reconnected at the main there is no requirement to upgrade. A majority of the subs I've checked had the neutarl bonded to the can. Been that way 30 years and stays that way.
From our standpoint the job was to change out the service panel, and that is what we inspect.

 

[Dennis Alwon]

To be honest I think they use to allow that around here. My concern, as an EC, is that I am creating an unsafe installation by upgrading if I don't rewire the sub panel.

To me that is part of the job. If they want an upgrade then they have to deal with the entire cost to do it right. If they can find someone else to do a half job then that's what they need to do.

I am not sure how you can allow it. I understand that the upgrade is for the service but by allowing the upgrade that way you are allowing a NEC violation that was not there before. I am assuming the sub panel was the original main panel.

Once you change the service and make it a sub then I believe you are responsibile to up grade the feeder to make it code compliant.

Suppose the range and dryer are fed from that main panel and have seu cable feeding their respective loads. Now what???? you got a mess.

 

[augie47]

upgrade

upgrade

Dennis, I don't disagree. I'm only relating a long standing policy in this area.
The original panel used the grounded conductor as the grounding conductor also...either because it was a service panel, or it was wired that way originally as a sub-panel (I don't recall when the N-G seperation rule became code). We approach an upgrade as an improvement. For us, to require subsequent steps often prevents the originl improvement to be completed (cost wise), or encourages folks to avoid inspections. If we have to change the existing feeder, do we then have to add grounding receptacles, ground faults, arc-faults , etc.? If I have $1,000 to replace an old fusible service panel in poor condition, must I wait until I have enough funds to upgrade other older wiring. We strongly encourage additional work, but are proud to see any steps to improve safety.

 

[Dennis Alwon]

While I understand your point, I don't agree with the analogy of replacing all the 2 wire circuits in the house. Leaving the receptacles as 2 wire does not create a violation.

If the panel you are talking about was already existing as an illegal sub panel, then I would agree with you but not if you create an illegal sub panel when you upgrade.

 

[al hildenbrand]

. . .if you create an illegal sub panel when you upgrade.

I think this is the crux of the matter. There is room in this for some subjectivity.

If I come to a dwelling with a single small service center and install a new service center in a different location, I can make the small old panel a subpanel by installing a feeder to it. The feeder doesn't exist until I install it. It has to have the EGC separate from the neutral.

If I come to a dwelling with a subpanel already supplied by a feeder from a remote service center then I need to take stock of the clues that tell me the approximate Code that applied. The classic is a K&T 30 Amp 240 / 120 Volt plug fuse center (as subpanel) in it's original location on the first floor on up to the attic that has been fed from the first electrical service upgrade, a 60 Amp 240 / 120 Volt 4 plug & 2 pull box, down in the basement.

When I replace that 60 Amp box with a new larger circuit breaker service center and service entrance, I am not installing a new feeder to the 30 Amp K&T subpanel. The feeder already exists. Like Augie, in my area, I reconnect the line end of the existing feeder in the new service that is replacing an existing service.

 

[George Stolz]

Al, good to see you back.

I hope Ed comes back as well to fill us in on the particulars of this installation.

 

[al hildenbrand]

:smile: Thanks George!

Meat space distracted me for a while there.

I, too, hope Ed chimes back with some more info.

 

[LarryFine]

If I come to a dwelling with a single small service center and install a new service center in a different location, I can make the small old panel a subpanel by installing a feeder to it. The feeder doesn't exist until I install it. It has to have the EGC separate from the neutral.

Not only that, but if there is an existing 3-wire feed to an appliance from said panel, you are supposed to re-wire it with a 4-wire feed and replace the receptacle and power cord.

Pre-existing does not always mean it can be left as found. Usually, yes, but not always.

Meat space distracted me for a while there.

Do we want to know? :-?

 

[al hildenbrand]

. . . if there is an existing 3-wire feed to an appliance from said panel, you are supposed to re-wire it with a 4-wire feed and replace the receptacle and power cord.

Well, the "supposed to" is my point. That's the subjective part. The AHJ and the history of enforcement for an occupancy's location will determine whether there is a "supposed to".

That 3-wire dryer or range receptacle, that exists out of the 60 Amp fuse box still exists after the replacement circuit breaker center goes in. It exists just like the general lighting branch circuits that one reconnnects to the new CB panel. These branch circuits are not new, therefore, in many jurisdictions, the current NEC doesn't extend out into the existing branch circuits, only to "new" work.

Do we want to know?

Nothing special, just life as usual amped up a little.

 

[ecirplr]

I am inspecting the electrical system of an older home, of which the owner does not necessarily want to re-wire, but nevertheless I want to provide earthing and bonding to make the electrical system as safe as possible. This in an older Victorian-era home with many rooms. The distribution consists of two main panels, and a system of 5 sub panels throughout the premises.

I am designing a good grounding system in the main distribution section following the guidlines of the NEC. It is the sub panels that I have some confusion on where to go.

Are you the AHJ and if so why are you "designing a good grounding system"? I have always been with the understanding that Inspectors did that and not "design". am not implying that Inspectors can't or shouldn't, but offer their expertise on the subject. It is relevant however that you have an understanding of the NEC and other applicable codes to substantuate your decisions.

The sub panels throughout this home are fed with 3 wires (2 hot conductors, and a neutral). No third grounding conductor was pulled. If there is a fault imposed on the chasis of one of these panels (or any other apparatus supplied by these panels), there is no path to clear this fault back to the source. I realize that the simplest way to solve this dilemma is to run a ground conductor from these panels back to the established ground at the main distribution system, but this may not be practical due to the architecture of the house.

My question is: Can I bond the neutral to the case of the sub panels to establish a path to ground? I realize that this is not acceptable under normal circumstanses, but in this case I cannot find any other method of grounding the equipment in this facility. Are there exceptions to the code in this circumstance?

Ed Battle

I have been in the trade for 30 years and a third generation electrician. That doesn't mean I know it all, but I do have an understanding of what a makes a "system safe". You seemed to haveanswered your own question with" not acceptable under normal conditions". There seems to always be an exception allowing loop holes for every situation. However when life and property are at stake I would go with what the NEC or local codes stipulate. You can't replace a life or make up lost time for rehabilitation due electrcal accidents. Do it right.