Subpanel Grounding

Status
Not open for further replies.
J

jeff43222

Senior Member
Location
Minneapolis, Minnesota
A friend of mine called me the other day asking about installing a subpanel in his pole barn. He lives so far out in the sticks that there is no AHJ regarding licensing, permits, inspections, etc. It's the kind of place where people tend to DIY their own electrical work (very badly, from what I've seen). Anyway, he asked me if he could run two hots and a neutral to the subpanel and ignore the EGC and instead just drive a ground rod at the pole barn. Naturally, I told him that was not code-compliant, but I had a hard time explaining why.

Of course, now I'm curious. Suppose he did the installation as described above. From a physics standpoint, why is it bad?
 
Re: Subpanel Grounding

jeff43222 said:
A friend of mine called me the other day asking about installing a subpanel in his pole barn. He lives so far out in the sticks that there is no AHJ regarding licensing, permits, inspections, etc. It's the kind of place where people tend to DIY their own electrical work (very badly, from what I've seen). Anyway, he asked me if he could run two hots and a neutral to the subpanel and ignore the EGC and instead just drive a ground rod at the pole barn. Naturally, I told him that was not code-compliant, but I had a hard time explaining why.

Of course, now I'm curious. Suppose he did the installation as described above. From a physics standpoint, why is it bad?
Click to expand...

It might well be code compliant.

It might even be safe, if there is no other conductive path from where he runs his circuit from out to the barn, such as a phone wire or water pipe.

The unsafe part is when you start to have current flow on the EGC or other conductive parts.

Personally, i wish they would amend the code to prohibit this kind of install.
 
As Bob said it may be code compliant if it falls under the 250.32(B)(2). Either way it will require connection to a grounding electrode.
 
Re: Subpanel Grounding

jeff43222 said:
Anyway, he asked me if he could run two hots and a neutral to the subpanel and ignore the EGC and instead just drive a ground rod at the pole barn. Naturally, I told him that was not code-compliant, but I had a hard time explaining why.
Click to expand...


Good old 250.32(B)2 is why you're having a hard time explaining it. :)
 
Re: Subpanel Grounding

petersonra said:
Personally, i wish they would amend the code to prohibit this kind of install.
Click to expand...

I'd be satisfied with making it a new-installation violation, but existing installs are permissable to use (or re-use).
 
Re: Subpanel Grounding

jeff43222 said:
Of course, now I'm curious. Suppose he did the installation as described above. From a physics standpoint, why is it bad?
Click to expand...
Jeff, have you missed out on all the good detached structure discussions?

Here's Mike Holt's answer to your question.

The constrasting argument is, if this setup is dangerous, then every service in the US is dangerous.
 
I often ponder this scenario also. With the three wire install, even with phone or cable servicing the detached building, for a dangerous parallel path to exist, the phone or cable line would have to be bonded to the grounded conductor of the feeder, correct? If most detached buildings are the way I see them here, the phone and cable are bonded at the main service entrance. From that point, the lines are just run out to where ever the phone or cable jacks are needed. Now the metallic water pipe is another story as we must use all grounding electrodes present so that one is a no brainer.
 
Re: Subpanel Grounding

jeff43222 said:
Suppose he did the installation as described above. From a physics standpoint, why is it bad?
Click to expand...
It's not bad, because its not dangerous. However, that relies on there really being no metal paths between the house and the barn, and no such paths ever being constructed in the future.

The physics of what happens in the barn is the same as in the house. You bond the N-G at the main panel, and at no other panel downstream. From any load anywhere in the house, the neutral current will travel along the neutral wire to the source. No current will travel along the EGC (or along any conduit, etc.) unless and until there is a fault within some piece of equipment. When the fault happens, the fault current travels along the EGC to the N-G bond, and from there to the source. This quickly trips the breaker. If you happened to be touching the tool when it has its failure, you may receive a shock, but it will be brief.

If the barn has no EGC in its feeder, and if the N-G is bonded at the barn's first panel (and at no other panel in the barn), then any neutral current from any load in the barn will travel along the neutral of the branch circuit to the barn panel, and along the neutral of the feeder to the source. That does not create a hazard to anyone operating a tool inside the barn. However, if a tool does experience a failure, then the fault current will travel along the EGC within the barn to the N-G bond at the barn panel, and from there along the neutral of the barn's feeder to the neutral of the main house panel. This will trip the breaker, and the person holding the tool may or may not feel a short shock.

The danger would come about if there were metal pipes between the buildings (or if metal pipes get installed at a future date). In that case, even under normal operation with no faulted equipment, neutral current from a tool inside the barn would travel along the neutral wire of the branch circuit to the barn's N-G bond, and from there along BOTH the neutral of the barn's feeder AND the metal pipe back to the house. This would energize the metal pipes (and other external metal parts of electrical equipment) within the house.
 
Even if current is being carried on the pipes in parallel to the neutral, what is the direct hazard? To me, this is no less harardous than touching the engine ground strap on a car while starting it -- there are hundreds of amps going through that wire, but the voltage is low.


The hazard I see is if the feeder neutral fails. If that happens, the pipes become the sole neutral and no one knows there's a problem. Then disconnect or cut the pipe one day when a load in the outbuilding is connected, and you could find yourself with 120V in one hand and a good ground on the other. You'll probably get some good arcing on the pipe when cutting it as warning, but that may come too late.
 
One thing really troubles me in this thread.

I do not think it has been made clear that if we use 250.32(B)(2) and run just 'hots and a neutral' to a remote building that we must install a bonding jumper at the remote building connecting the new EGC created at the building to the neutral.
 
suemarkp said:
Even if current is being carried on the pipes in parallel to the neutral, what is the direct hazard?
Click to expand...

There is no hazard if we are talking about a service.

These parallel paths are everywhere on the line side of the service disconect.

Once we get on the load side of the service it is a hazard.


Or at least that seems to be the NFPAs opinion. :p :wink:
 
Status
Not open for further replies.
Top