300.20 Gone Wild

[don_resqcapt19]

So, you are saying that the rule language requires that a multiconductor cable with twisted conductors must maintain the conductor twist inside the enclosure of ferrous metal?

The very definition you have cited for "grouped" is fighting you in your interpretation. The definition of grouped says the conductors need only be adjacent AND NOT in continuous contact.

No where have I said they have to be twisted...I have only said that the code rule requires them to be grouped, and in my opinion one conductor running down the wireway on one side of the panel is not grouped with a conductor running down the wireway on the other side of the panel.

 

[don_resqcapt19]

There is no need for the last sentence in the code section. If it is to be changed you just delete the following words:
"To accomplish this, all phase conductors and, where used, the grounded conductor and all equipment grounding conductors shall be grouped together"

 

[ActionDave]

There is no need for the last sentence in the code section. If it is to be changed you just delete the following words:
"To accomplish this, all phase conductors and, where used, the grounded conductor and all equipment grounding conductors shall be grouped together"

I'll agree with that. I'm always in favour of deleting words from the code.

 

[iwire]

Has anyone, out on a job failed inspection due to wires not being grouped in a panel?

This is a ridiculous thread.

 

[romex jockey]

I've had "more than 24" pointed out , but never in a panel. Fact is, some of the larger Eframe panels have all the stranded wire tie wrapped for 60" or more....

~RJ~

 

[peter d]

Has anyone, out on a job failed inspection due to wires not being grouped in a panel?

This is a ridiculous thread.

:thumbsup:

Sometimes people like to argue just for the sake of arguing.

 

[romex jockey]

I'd rather be enlightened to the theory behind a code, than the code itself

But the devil is , as usual, in the details.....

~RJ~

 

[kwired]

I'd go a step further with the edit.



Get rid of "grouped".

I still want to see it say, in some fashion, that it's okay to run a switch leg that's not accompanied by the neutral. But I think that's for another thread.

I don't feel any change is needed to cover the unused neutral in a switch leg.

"To accomplish this, all phase conductors and, where used, the grounded conductor"

Bold part in current 2014 wording already gets the job done. If a neutral is not used you don't need to provide one. However you do have 404.2(C) that may require one anyway, or at least with raceways, leave enough "fill" space to add one if needed. Remember though that not all switches fit what is required in 404.2(C).

After some more thought I guess I can see people making "and where used" mean more then one thing, so maybe a little adjustment somehow to clarify intent wouldn't hurt.

 

[al hildenbrand]

I'd go a step further with the edit.

Get rid of "grouped".

I still want to see it say, in some fashion, that it's okay to run a switch leg that's not accompanied by the neutral. But I think that's for another thread.

I agree with Kwired's last post that there is the support of the switch loop sans neutral already in 300.20. The key is in the definition of the NFPA Glossary, that Don gave us, of "grouped".

No where have I said they have to be twisted...

Thank you, Don. Now, letting the OP's question rest as answered, by the Forum, IMO, I want to keep poking at "grouped" as you seem to be misreading the very Glossary definition that you provided for us.

I have only said that the code rule requires them to be grouped, and in my opinion one conductor running down the wireway on one side of the panel is not grouped with a conductor running down the wireway on the other side of the panel.

Don, I agree with you that 300.20 DOES tell us to not do that, but, because of the first sentence use of "arranged" so as to avoid heating the surrounding ferrous metal by induction." When two AC conductors of a single circuit are arranged as you describe in the bold red of your words, above, their magnetic fields are additive, not canceling, giving rise to induced currents in surrounding ferrous metal. This is a bad ARRANGEMENT giving rise to induction in surrounding ferrous metal.

The words of the first sentence of 300.20 do not give a threshold of heating by induction. I agree with you that this is all-encompassing, unlike the 200 Amp threshold that you report is in the Canadian Electrical Code. BUT, the rule hinges on "heating. . .by induction", NOT grouped. Your quote of the NFPA Glossary of Terms definition for "grouped" for all of NFPA 70 says unambiguously that grouped conductors are positioned ADJACENT to one another, AND grouped conductors are NOT in continuous contact with each other.

That NFPA Glossary definition greatly blurs the meaning away from the "singularity" that you appear to claim it has. Especially when you, Don, point to the NFPA Glossary definition of "adjacent" as not applying to grouped, which throws us to the Miriam Webster.

Don, you have been instrumental in answering Rob's OP related question:

I guess that we now need a clear definition of grouped. . .
Will anyone take a shot at defining the word grouped . . .?

There is no need for the last sentence in the code section. If it is to be changed you just delete the following words:
"To accomplish this, all phase conductors and, where used, the grounded conductor and all equipment grounding conductors shall be grouped together"

Including EGCs with CCCs I believe to be important, and removing that is problematic.

There is a real technical issue with how to measure "heating by induction" and to tease it out relative to ambient temps and I squared R heating in the current paths of conductors and terminations, as well as the heat created by all the solid state circuitry that is being placed inside the overcurrent protective devices.

 

[jaggedben]

Has anyone, out on a job failed inspection due to wires not being grouped in a panel?

Well, the OP, evidently.

I don't feel any change is needed to cover the unused neutral in a switch leg.

"To accomplish this, all phase conductors and, where used, the grounded conductor"

...

Please tell this to the inspectors who insisted I run a neutral 50 or 100ft to a disconnect where it does not terminate, then back to the T-condulet where the switch legs branched off. Mind you this was not for a lighting circuit, it was for a solar output where the neutral doesn't even carry current! And not only that, but the inspector said that he understood that and understood the physics but that we still had to do it anyway because that was the letter of the code!

No, for sure I wish the code would state this part explicitly.

 

[kwired]

Well, the OP, evidently.



Please tell this to the inspectors who insisted I run a neutral 50 or 100ft to a disconnect where it does not terminate, then back to the T-condulet where the switch legs branched off.

No, I wish the code would state this part explicitly.

I can tell them, but I can't make them listen

 

[don_resqcapt19]

...

That NFPA Glossary definition greatly blurs the meaning away from the "singularity" that you appear to claim it has. Especially when you, Don, point to the NFPA Glossary definition of "adjacent" as not applying to grouped, which throws us to the Miriam Webster.

The west lot line of my property is not adjacent to the east lot line of my property. The same applies to an enclosure....one side is not adjacent to the opposite side.

...
Including EGCs with CCCs I believe to be important, and removing that is problematic. ...

I don't really see an issue within an enclosure, and we have 300.3(B) to require the EGC be run with the circuit conductors.

As I said, I don't see a code intent to group within the enclosure, but I see code language that requires something other than the intent. However, there could be cases in large enclosures with very high currents, where there would be excessive heating if the conductors are not grouped, but in my opinion the first sentence of the section covers that....it just doesn't give a prescriptive method to prevent it.

Most of the NEC is prescriptive and it seems that they want to keep it that way, but there are a few "performance" rules and this could be come one of them if the last part was delted.

 

[al hildenbrand]

The west lot line of my property is not adjacent to the east lot line of my property. The same applies to an enclosure....one side is not adjacent to the opposite side.

You are dividing the border into parts of a border and then playing a word game. Your "lot line" is a border. The west part of the "lot line" is part of the border. That doesn't fit in the common definition, that is, unless you expand to look for a "border" around your "lot line" which could easily be the outer border of the "adjacent" LOTS to your lot. Relative to your east and west lot lines, this multi-lot border is common to them, and your east and west lot lines ARE adjacent.

Simple Definition of adjacent

1 : close or near : sharing a border, wall, or point​

Full Definition of adjacent

• 
  [*=1]
  a : not distant : nearby <the city and adjacent suburbs>​

b : having a common endpoint or border <adjacent lots> <adjacent sides of a triangle>​

c : immediately preceding or following​

2 : of two angles : having the vertex and one side in common​

​

 

[al hildenbrand]

Simple Definition of adjacent

1 : close or near : sharing a border, wall, or point​

Remember, NFPA Glossary of Terms states that the conductors are "not in continuous contact with one another". . . this doesn't say that they HAVE to be in contact at at least one point, which, to me, rules out the word "point" in the above definition leaving border or wall.

AND, since the entire second sentence of 300.20(A) is tied to the first sentence through the object of "this" in "To accomplish this " they absolutely work together, not isolated, as you claim. That means the common border is the ferrous metal subject to inductive heating.

 

[infinity]

I started this thread because the exact same information came to me from two completely different sources. One a master electrician and the company license holder and two from an apprentice school teacher. I was asked to weigh in with my opinion which was that it is bogus and not supported by the NEC because panels like the one wired in post #98 have been wired that way for a hundred years and do not cause any problems.

 

[al hildenbrand]

I can't recall an earlier discussion of 300.20(A) that included the tie in from the NFPA Glossary of Terms for "grouped". That definition, in my opinion, helps a lot.

 

[don_resqcapt19]

You are dividing the border into parts of a border and then playing a word game. Your "lot line" is a border. The west part of the "lot line" is part of the border. That doesn't fit in the common definition, that is, unless you expand to look for a "border" around your "lot line" which could easily be the outer border of the "adjacent" LOTS to your lot. Relative to your east and west lot lines, this multi-lot border is common to them, and your east and west lot lines ARE adjacent.

​

Al,
We are not going to agree on this issue...there is just no way that wires run down one wireway are grouped with the wires in the opposite wireway (my opinion and it is not going to change).

 

[don_resqcapt19]

I started this thread because the exact same information came to me from two completely different sources. One a master electrician and the company license holder and two from an apprentice school teacher. I was asked to weigh in with my opinion which was that it is bogus and not supported by the NEC because panels like the one wired in post #98 have been wired that way for a hundred years and do not cause any problems.

Rob,
I see that as an actual violation, but not an intended one, and not one that a reasonable inspector would cite...this is just poor code language. The fact that it is not a real world problem does nothing to change the code language.

 

[infinity]

Rob,
I see that as an actual violation, but not an intended one, and not one that a reasonable inspector would cite...this is just poor code language. The fact that it is not a real world problem does nothing to change the code language.

I tend to agree but anyone inspecting something like this should use some common sense and think about the actual reason that those words are in the NEC in the first place.

 

[Smart $]

I tend to agree but anyone inspecting something like this should use some common sense and think about the actual reason that those words are in the NEC in the first place.

That may actually be the problem.

The result is just not what we expected. :slaphead: