Bonding Multiple Service Disconnecting Means - Objectionable Current

[W6SJK]

Couldn't find a Mike Holt graphic for this - pardon me

In the scenario with 2 or more separate service disconnecting means, each with N-G bonding, we will have current flowing on the neutral and the enclosures / metallic raceways under normal and fault conditions, correct? Even worse if one disco is 5-10 ft away due to wall space. Never really noticed that since I always deal with service switchboards.

Cheers!

 

[infinity]

Yes if the nipples are metal then they will be a parallel path for the neutral current. One of the few places where the NEC allows objectionable current.

 

[LarryFine]

I don't think of it as objectionable current; rather, I see the service neutral(s) and the metallic enclosures and raceways as one large, single, zero-volts reference point for the premises.

That's why the service neutral conductor can be bare. Even though it carries current, the customer's end of the drop/lateral's neutral is re-established as a grounded supply conductor.

 

[wwhitney]

That current on the service enclosures isn't objectionable, but what is objectionable is that with the 3 service disconnects in the picture, with 3 different MBJs you get 3 different EGC systems. And if those EGC systems are every interconnected downstream (as would be common and perhaps expected, e.g. the same metal water pipe serves two pieces of equipment, each of which is supplied from a different service disconnect), now you have an alternate path for neutral current. Namely MBJ1 - EGC1 - EGC interconnect - EGC2 - MBJ2. So some neutral current will take that path and put neutral current on the EGC systems.

It seems to me that we would be better off if the NEC were to allow a single MBJ in the trough, rather than having 3 separate MBJs. Failing that, it seems like it would be a good idea to immediately bond together the 3 EGC systems at a common point next to all the enclosures. That way the above objectionable current is limited to between the common bond point and the MBJs.

Cheers, Wayne

 

[LarryFine]

It seems to me that we would be better off if the NEC were to allow a single MBJ in the trough, rather than having 3 separate MBJs.

I thought it does allow that.

 

[jaggedben]

Yes if the nipples are metal then they will be a parallel path for the neutral current. One of the few places where the NEC allows objectionable current.

Or, it's not objectionable because the NEC permits it.

 

[jaggedben]

I thought it does allow that.

It effectively doesn't because 250.24(B) requires an MBJ in each service disconnect.

 

[jaggedben]

... Failing that, it seems like it would be a good idea to immediately bond together the 3 EGC systems at a common point next to all the enclosures. That way the above objectionable current is limited to between the common bond point and the MBJs.

The common neutral bar in the trough already effectively does that (as do the trough and raceways connection the disconnects, if metal and properly bonded).

I do appreciate your overall point, and I suppose there could be an outlier case where one or more service disconnects are remote and the resulting parallel EGC path is the same length as the proper neutral path.

 

[LarryFine]

Or, it's not objectionable because the NEC permits it.

Requires it.

 

[LarryFine]

It effectively doesn't because 250.24(B) requires an MBJ in each service disconnect.

Right.

You're referring to bonding the disconnect enclosures to the neutral.

I was picturing connecting the grounding electrodes to the neutral.

 

[wwhitney]

The common neutral bar in the trough already effectively does that (as do the trough and raceways connection the disconnects, if metal and properly bonded).

But the common grounded conductor bar (or common grounded service conductor if there isn't a common neutral bar) is on the utility side of the 3 MBJs. So there will be some conductor length, and hence resistance, between each MBJ and that common point. That means that some current will take any available loop path through the EGCs.

If we provide a load side common EGC point that all the EGCs pass through, then any EGC loop downstream of that common EGC point should have zero grounded conductor current through it. As that common point will have a single potential associated with it, and there will be no potential difference to drive current around a downstream loop.

Cheers, Wayne