225.39(D) Where is feeder disconnect permitted?

[gene6]

Wyane I appreciate the help, I really do.
But this CMP stuff while not new to me, I have not payed close attention to it.
I know if they like part of a proposal they can 'accept in part'. I am told 2002 was the cycle after this was added, my boss has a 1996 book and its not in there.
So the question comes up and got knocked down and people notice. Why ask again when you know the answer?

That code change proposal was for over-current protection nothing to do with the "branch circuit or feeder disconnect".
Per the definition of a feeder there needs to be a disconnect at the beginning of the feeder.
Where else is a "branch circuit or feeder disconnect" ?
This fella is a long time inspector and clearly stated his argument.
If a "branch circuit or feeder disconnect" is not different from a "building or structure disconnect" and "the disconnecting means specified in 225.31" why word it differently just that one time in 225.39?
Why not say like they did in 225.36 "The disconnecting means specified in 225.31" ?

 

[Little Bill]

A feeder is a feed to a disconnect with OCPD. Code says to put the disconnect nearest the point of entry to the building/structure, so how could the disconnect be anywhere but where the feeder ends? The panel feeding it is the source. Although you can turn it on/off from there, it is not the disconnect to the separate structure. A fireman wouldn't want to have to go all the way back to the source to kill power at a remote structure. Nor would anyone in the structure want to go back when they can disconnect it right at/in the structure.

 

[wwhitney]

Per the definition of a feeder there needs to be a disconnect at the beginning of the feeder.

No, there is no requirement for a disconnect at the beginning of a feeder. Just for an overcurrent device.

This fella is a long time inspector and clearly stated his argument.

But this inspector is confusing a disconnect requirement with an overcurrent device requirement. A disconnect and an overcurrent device are two different things in principle. No section in the NEC that wishes to regulate the size of an OCPD refers to it as a disconnecting means.

If a "branch circuit or feeder disconnect" is not different from a "building or structure disconnect" and "the disconnecting means specified in 225.31" why word it differently just that one time in 225.39?

Inertia? Imprecision? That's a better explanation than that they want to subtly regulate the OCPD at the supply end of the feeder, without ever referring to the term "overcurrent device".

Why not say like they did in 225.36 "The disconnecting means specified in 225.31" ?

PI submitted.

BTW, Mike Holt submitted a PI for 225.39 for the 2023 NEC which would have added the phrase "conductors and" in front of "disconnecting means" whenever the rating was discussed in 225.39. The panel rejected it with the statement "The requirement for conductors to be sized per the calculated load is already referenced in Articles 210, and 215. There's no need to add the proposed language to 225.39."

So clearly the conductors are only required to be sized to the calculated load, not to the minimum sizes listed in 225.39. Likewise, the OCPD at the source end (otherwise you would be required to upsize the conductors).

Cheers, Wayne

 

[gene6]

No, there is no requirement for a disconnect at the beginning of a feeder. Just for an overcurrent device.

Please provide an example of a feeder with only fuses and no disconnect that meets code.
See 240.40 and 240.21

 

[jaggedben]

Please provide an example of a feeder with only fuses and no disconnect that meets code.
See 240.40 and 240.21

100A service panel with 100A main breaker and feed through lugs.
#2 wire goes from feed through lugs to a fuse block with 30A fuses.
#10 wire goes to a panel with breakers

While you'd likely never see this on the last 100 years, because who is going to source a fuse block instead of a fused disco, it's still technically code compliant as far as I know. 215 contains no disconnecting means section.

 

[wwhitney]

Please provide an example of a feeder with only fuses and no disconnect that meets code.
See 240.40 and 240.21

See the previous post. The point is that 240 is about overcurrent devices, not about disconnects (with the exception of 240.40).

Your inspector's mistaken argument at best implies that if a feeder from building A to building B has fuses as its overcurrent device at building A, then 225.39 regulates the disconnect at building A, rather than the disconnect at building B. I.e. if building B's load is 30A, but 225.39 requires a 60A disconnect, you would need a 60A disconnect at building A, but could use 30A fuses at building A, and a 30A disconnect at building B.

The weight of everything said in this thread shows that is wrong. 225.39 is regulating the disconnect at building B, not at building A. So in this example, you could have a 30A disconnect at building A, 30A fuses at building A, and would need a 60A disconnect at building B.

Cheers, Wayne

 

[gene6]

100A service panel with 100A main breaker and feed through lugs.
#2 wire goes from feed through lugs to a fuse block with 30A fuses.
#10 wire goes to a panel with breakers

While you'd likely never see this on the last 100 years, because who is going to source a fuse block instead of a fused disco, it's still technically code compliant as far as I know. 215 contains no disconnecting means section.

That would violate the "each circuit containing fuses can be independently disconnected
from the source of power." part of 240.40

225.39 is regulating the disconnect at building B, not at building A.

If someone was digging a ditch near the feeder and I asked you to throw the 'feeder disconnect' for safety which one would you throw the one at A or B?
Now re-read 225.39 and apply Charlies rules.

 

[jaggedben]

If someone was digging a ditch near the feeder and I asked you to throw the 'feeder disconnect' for safety which one would you throw the one at A or B?

Ask me which one I'd throw if someone was working in the outbuilding instead of digging a ditch. For good measure, pretend it's a 500ft walk to building A and someone is impatient to get started.

(You've begged the question.)

Now re-read 225.39 and apply Charlies rules.

There's two disconnecting means. How do Charlies rules help you figure out which one it's referring to?

 

[gene6]

There's two disconnecting means. How do Charlies rules help you figure out which one it's referring to?

Ben because 225.39 says what it says not what you think it says, not what other paragraphs near by says, just what it says "feeder disconnect" not "building or structure disconnect".
The same disconnect you would throw if you were working on the feeder.

Charlies's Rule for Reading the NEC

This is the often mentioned Charlies Rule. Charlie is one of our long time moderators.

forums.mikeholt.com

 

[jaggedben]

No, there's literally two feeder disconnects. (Note that the feeder at the outbuilding could travel some distance after the disconnect.)

Say we are on a jobsite, and the feeder goes from building A to a disconnect on the outside of Building B, and then continues on to a panel inside Building B. If you ask me to throw the feeder disconnect for Building B, I'm going to ask you 'Which one?"

 

[wwhitney]

Ben because 225.39 says what it says not what you think it says, not what other paragraphs near by says, just what it says "feeder disconnect" not "building or structure disconnect".
The same disconnect you would throw if you were working on the feeder.

But from the context of Part II of Article 225, and the subject of Sections 225.31 through 225.38, it is clear that 225.39 is referring to the disconnect at the building, as specified in 225.31. It is not plausible that Part II of Article 225 would jump from talking about one disconnect to a totally different disconnect without being much more explicit.

You and your inspector are just misinterpreting the slight change in terminology in 225.39. The magnitude of that terminological change is much smaller than the magnitude of the semantic difference you are reading into it. And the CMP responses to PIs I have referenced demonstrate that the CMP responsible for 225.39 agrees.

Cheers, Wayne