Trying to Clarify Feeder, Branch, and Tap Terms in Industrial Control Panels

[Shackled Designer]

The background for the questions below is industrial panel design involving a primary panel that will supply at least one subpanel.

1) Is it correct to say that a supply line from the main panel to a subpanel is considered a feeder from the perspective of that subpanel?

2) If within the main panel I protect the supply to the subpanel with a branch circuit protector, does that mean that the subpanel supply line is not considered to be a tap (and therefore not subject to tap rules)?

3) If the supply to the subpanel is not a tap as I have described it in question 2, are there any other NEC considerations for the wire run between panels that I need to bear in mind beyond sufficient OCP for the subpanel's load?

Kind regards,
Shak

 

[GoldDigger]

1. It is a feeder period. The definition applies.
2. Yes, any conductor protected upstream to its ampacity and load would not need to conform to tap rules.
3. Unless it is part of a PV or other backfeed, none come to mind.
Which is to say that there are a lot of NEC requirements for wiring methods, but none which are particular to this situation.

 

[augie47]

It might help to take the Art 240 definition of a tap into consideration:
Tap Conductors.
As used in this article, a tap conductor is defined as a conductor, other than a service conductor, that has overcurrent protection ahead of its point of supply that exceeds the value permitted for similar conductors that are protected as described elsewhere in 240.4 .

Your conductor must be adequate to handle the load and must be protected as noted in Art 240.4

 

[Shackled Designer]

GoldDigger,

Thank you for your answers. Regarding #3, no there are no photo-voltaics or other backfeeds involved.


augie47,

I did examine the tap definition in 240, and it appeared to answer my second question. I wanted to verify my interpretation of the tap definition, however, since the definition does not explicitly refer to the tap conductors supplying other panels. Then again, it does seem more apparent from the tap rules (10 ft. taps, 25 ft. taps, etc.), that conductors to something outside the source panel is in mind. Come to think of it, by the definition, does any circuit that does not exit the panel qualify as a tap?

Maybe taps are used because a branch-to-subpanel circuit is more expensive (owing to the presence of the BCPD in the source panel). On the other hand, even taps have some OCP requirements. It looks like there are other hurdles to be jumped in order to use taps, and this makes the branch-to-subpanel approach seem more appealing to me. I would be interested to know which approach others prefer and why.


Thanks for whatever further comments are made. I appreciate the learning.

Best regards,
Shak

 

[starbolin]

GoldDigger,

Thank you for your answers. Regarding #3, no there are no photo-voltaics or other backfeeds involved.


augie47,

I did examine the tap definition in 240, and it appeared to answer my second question. I wanted to verify my interpretation of the tap definition, however, since the definition does not explicitly refer to the tap conductors supplying other panels.

Does not depend on whether the conductors supply a panel or not.

Then again, it does seem more apparent from the tap rules (10 ft. taps, 25 ft. taps, etc.), that conductors to something outside the source panel is in mind.

I don't know 'source panel'. Can be within the same panel or not.

Come to think of it, by the definition, does any circuit that does not exit the panel qualify as a tap?

Not sure what you meant by 'any circuit'. Conductors between the service and the final OCPD are a 'feeder' ( Article 100, Definitions, 'Feeder'. Feeder conductors with ampacity less than their OCPD ratings are a 'tap' 240.2, Definitions, 'Tap Conductors'.

Note: all taps are feeders but not all feeders are taps.

Maybe taps are used because a branch-to-subpanel circuit is more expensive (owing to the presence of the BCPD in the source panel).

Expense is one reason. There are others. It's just a thing that is permitted. A tool for the designer/engineer/installer to use. The code is not prescriptive. It's not a template. The code is permissive. It merely states what is allowed and what is not.

On the other hand, even taps have some OCP requirements.

Feeders have OCP requirements. Taps are an allowance for feeders that allow a reduced conductor ampacity in areas with sufficient physical protection.

It looks like there are other hurdles to be jumped in order to use taps, and this makes the branch-to-subpanel approach seem more appealing to me.

That decision is why you get the big bucks.

I would be interested to know which approach others prefer and why.

It's not a matter of what I 'prefer'. I bring the tool to the job that the problem required. In general my electrical work doesn't require 'value-engineering' so I'm not going to make my life harder by trying to shoe-horn in a long tap. But, were I to have a situation where I had multiple disconnects in a large enclosure or coming off of a wire-way then, yes, I am going to use the tap rules. I can't even begin to enumerate all the ifs-ands-and-buts in those two statements so I am not even going to try.

 

[Shackled Designer]

starbolin,

Thank you for your detailed dissection of my question.

So if I have a panel protected by a 150A fusible disconnect, and the feeder circuit contains power distribution blocks for multiple branch circuits, then by your understanding of tap conductors would you say that any load conductors from the PDBs are taps if they are smaller than 1/0? (Figure a 75?C temp rating on the conductor insulation.)

Or would this be me applying the definition of the tap conductor to something for which it was not intended?

Thanks,
Shak

 

[starbolin]

starbolin,

Thank you for your detailed dissection of my question.

So if I have a panel protected by a 150A fusible disconnect, and the feeder circuit contains power distribution blocks for multiple branch circuits, then by your understanding of tap conductors would you say that any load conductors from the PDBs are taps if they are smaller than 1/0? (Figure a 75?C temp rating on the conductor insulation.)

Or would this be me applying the definition of the tap conductor to something for which it was not intended?

Thanks,
Shak

Sorry, I got a bit confused. Is there a downstream OCPD? Everything upstream from the last OCPD to the service is a feeder. Everything downstream from the last OCPD to the load is a branch. Branch circuit taps are 240.21(A) and feeder taps are 240.21(B). Which are we discussing?

 

[Shackled Designer]

If we are talking about taps at all, I guess that they would be feeder taps.

After the power distribution blocks that I mentioned, the load-side conductors from each block terminate at the BCPs of each branch.

Part of what is throwing me here is that if these conductors are taps, then the feeder tap rules require that I run these through raceway to their destinations. But if the first couple of BCPs are located within a few inches of the feeder distribution, then cutting a 2-inch piece of raceway to meet code sounds silly.

Alternatively, if, in the 150A panel example I previously gave, the tap definition applies to any wire smaller than 1/0, then I maybe could select PDBs which support the 1/0 wire gauge on their load side, but feeding such a large wire into a branch with, say, a 6A fuse block protecting an AC-DC power supply would also seem pretty goofy.

Although the tap definition does not qualify whether a conductor is considered a tap on the basis of whether it exits the enclosure (panel) from which it originates, the rules all seem to suggest that taps exiting an enclosure was why that section of code was written.

Trying to do it all the right way,
Shak

 

[GoldDigger]

If they do not exit the panel, they are arguably part of the UL listed panel assembly and not under the jurisdiction of the NEC at all, at least not in terms of being taps.
Even if the wires totally inside the panel are field installed wires, at least they are protected inside the panel and are less likely to start a fire elsewhere in the building.

 

[kwired]

The art 240 definition of "Tap Conductor" only applies to a conductor that has overcurrent protection that is higher then typically permitted on the supply side of said conductor. Does not matter if that conductor originates in a panel, disconnecting means, splice box etc. It does not apply to any conductor that is a part of a listed assembly, that conductor has been evaluated and is deemed acceptable for the assembly.

You also seem to be confusing feeders and branch circuits to some extent. Go to art 100 and see the following definitions:

Branch Circuit.
The circuit conductors between the final overcurrent device protecting the circuit and the outlet(s).

Feeder.
All circuit conductors between the service equipment, the source of a separately derived system, or other power supply source and the final branch-circuit overcurrent device.

A branch circuit will supply a load, or at least an "outlet" where a load will ultimately connect to somehow, a feeder may supply additional feeders or a branch circuit or multiple branch circuits.

Both have overcurrent protection rules that generally require conductors to be protected at or below the ampacity of the conductor (some exceptions for certain equipment) If either has overcurrent protection that is higher then the ampacity of the conductor (outside of certain exceptions allowed for certain equipment), then you have a art 240 volt "tap conductor", it will either be a branch circuit tap conductor or a feeder tap conductor, or a secondary conductor of a transformer (which is kind of same as a feeder tap, but they do address it separately)


Feeder tap conductors seem to get talked about the most often followed by transformer secondary conductors - those are more complicated and there is more common applications where those occur then for branch circuit tap conductors.

If the word tap is used anywhere else outside art 240 or you are not talking about overcurent protection it doesn't necessarily have the same meaning.

If you bring a main conductor that is a feeder or branch circuit conductor into an enclosure, and within enclosure you split into multiple directions via any method and leave the splice box with a conductor that is still provided with proper overcurrent protection for it's ampacity - you do not have an art 240 tap conductor.

In your example of a 150 amp panel you say anything leaving that is smaller then 1/0 is a tap - only true if the conductor has 150 amp overcurrent protection on it's supply side, should it have a fuse or breaker protecting it at or below it's ampacity and on the supply side it is not a tap. Now if you have a short conductor from the 150 amp protected conductor to the overcurrent device it is a tap. If it doesn't leave the enclosure where the tap is made - you have different rules then for a tap that does have some distance to it.

 

[Shackled Designer]

GoldDigger, kwired,

I was not aware that a listed assembly falls outside the purview of the NEC. Thank you for educating me on this point. Even so, not all the panels with which I am involved are reviewed by UL, and so I had better be well-informed for both cases.


As far as the definitions of feeders and branches, the definitions seem very simple when one deals with a single panel. However, to the extent that my main panel "feeds" subpanels with power components in them, I have to evaluate the SCCR for each power panel independently. The SCCR determination (per UL508A, supplement SB), as I understand it, relies on what are branch and feeder circuits from the perspective of that panel. I asked questions #1 and #2 in my original post not only to get tap conductors straight in my thinking, but also to verify that a branch in one panel could act as the feeder to another panel. Is this where I confused the two?

A branch circuit will supply a load, or at least an "outlet" where a load will ultimately connect to somehow, a feeder may supply additional feeders or a branch circuit or multiple branch circuits.

So, are you saying that the supply line that feeds the subpanel from the main panel is not a branch but continues to be a feeder? If that is the case, then putting a BCPD on that line in the main panel does not make it a branch so long as I have another "final" BCPD in the subpanel. Please tell me if I still misunderstand.

In case it matters, the reason that I would consider putting a BCPD in the main cabinet as well as in the subpanel is to reduce the size of the power wires running between the two. Of course, I could leverage the tap conductor rules provided my external wire runs are short enough.

Since I made my original post, however, the information that has been so far presented has changed my question about tap conductors.

Now if you have a short conductor from the 150 amp protected conductor to the overcurrent device it is a tap. If it doesn't leave the enclosure where the tap is made - you have different rules then for a tap that does have some distance to it.

This statement answers the question about whether the conductors have to exit the panel enclosure in order to be taps (they do not), but the new question it creates is this: where or what are these "different rules" for the short run taps that do not leave the enclosure? As I have already indicated, the rules in article 240.21 seem to apply only to conductors which do exit the panel, or do I misapprehend these as well?

Thank you for your patience and persistence on this.
Shak

 

[GoldDigger]

You should not read to much into the term BCPD, since it could just as well be used on a feeder

Try not to classify all of the non-main breakers in the main panel as branch breakers. Some may be feeders and you cannot tell by looking at them which they are. (You may make some guesses based on size, but those guesses could be wrong.
It is the wiring downstream of the breaker that determines what it is.
You could have a service for which there is exactly one load. I'm that case the service disconnect could be a branch breaker.

 

[kwired]

GoldDigger, kwired,

I was not aware that a listed assembly falls outside the purview of the NEC. Thank you for educating me on this point. Even so, not all the panels with which I am involved are reviewed by UL, and so I had better be well-informed for both cases.

If what you are building is listed, you build it to the applicable UL standards and not the NEC. You can factor in the fact that the installer likely needs to follow NEC and possibly make some things easier for the installer to comply with NEC. But things like overcurrent protection of a conductor that is a part of the listed equipment are things that are regulated by the listing standard and not necessarily the NEC.

As far as the definitions of feeders and branches, the definitions seem very simple when one deals with a single panel. However, to the extent that my main panel "feeds" subpanels with power components in them, I have to evaluate the SCCR for each power panel independently. The SCCR determination (per UL508A, supplement SB), as I understand it, relies on what are branch and feeder circuits from the perspective of that panel. I asked questions #1 and #2 in my original post not only to get tap conductors straight in my thinking, but also to verify that a branch in one panel could act as the feeder to another panel. Is this where I confused the two?

A "branch" from one panel can be a feeder. "Branch circuit" is a defined term in art 100. A panel can supply feeders or branch circuits. A feeder or branch circuit does not have to come from a "panel", feeders/branch circuits are determined by overcurrent protection arrangements and what is at the load end.

So, are you saying that the supply line that feeds the subpanel from the main panel is not a branch but continues to be a feeder? If that is the case, then putting a BCPD on that line in the main panel does not make it a branch so long as I have another "final" BCPD in the subpanel. Please tell me if I still misunderstand.

I think you are getting it, but as I said rules could be different if building listed equipment, but that is pretty much the way it is if this is an NEC application. Any field installed equipment between two listed items, especially if not supplied or specified by the listed equipment usually NEC governed.

In case it matters, the reason that I would consider putting a BCPD in the main cabinet as well as in the subpanel is to reduce the size of the power wires running between the two. Of course, I could leverage the tap conductor rules provided my external wire runs are short enough.

That would be the general rule for NEC applications. If you are supplying certain loads - motors being a common application, you may have a small conductor in relation to the overcurrent device, but this is a modification to the general rules and is addressed in art 430 for motors.

This statement answers the question about whether the conductors have to exit the panel enclosure in order to be taps (they do not), but the new question it creates is this: where or what are these "different rules" for the short run taps that do not leave the enclosure? As I have already indicated, the rules in article 240.21 seem to apply only to conductors which do exit the panel, or do I misapprehend these as well?

I maybe misled you slightly about rules for taps that don't leave the enclosure where the tap is made, there really is only one rule and it is in 240.21(B)(1)(4) and it reads:

For field installations, if the tap conductors leave the enclosure or vault in which the tap is made, the ampacity of the tap conductors is not less than one-tenth of the rating of the overcurrent device protecting the feeder conductors.

Note it says for field installations - so it would not apply to construction of listed equipment, any limitations for construction of listed equipment would be covered by the listing requirements.

 

[Shackled Designer]

" . . . the code is more what you'd call 'guidelines' than actual rules." -- from Pirates of the Caribbean: The Curse of the Black Pearl

OK. That statement doesn't apply to the NEC . . . but I can take a moment of pleasure in imagining . . .


Thank you, kwired, and GoldDigger, for your further input.

I maybe misled you slightly about rules for taps that don't leave the enclosure where the tap is made, there really is only one rule and it is in 240.21(B)(1)(4) and it reads:

For field installations, if the tap conductors leave the enclosure or vault in which the tap is made, the ampacity of the tap conductors is not less than one-tenth of the rating of the overcurrent device protecting the feeder conductors.

Note it says for field installations - so it would not apply to construction of listed equipment, any limitations for construction of listed equipment would be covered by the listing requirements.

So it sounds like for the scenario I have indicated (a set of feeder tap conductors terminating at a one of the BCPs only several inches from the tap point of origin and in the same panel), if the panel is not a listed assembly, then the NEC rules that apply are 240.21 (B)(1)(1), regarding conductor ampacity; 240.21 (B)(1)(2), pertaining to the limiting of the tap conductors to the device that the tap is meant to supply; and 240.21 (B)(1)(3), which states that raceway must be used from the tap to another enclosure . . . "Except at the point of connection to the feeder . . ."

Provided that this exception applies to the short runs that I have described, then maybe my angst over all this is much ado about nothing.

Best regards,
Shak

 

[kwired]

" . . . the code is more what you'd call 'guidelines' than actual rules." -- from Pirates of the Caribbean: The Curse of the Black Pearl

OK. That statement doesn't apply to the NEC . . . but I can take a moment of pleasure in imagining . . .


Thank you, kwired, and GoldDigger, for your further input.



So it sounds like for the scenario I have indicated (a set of feeder tap conductors terminating at a one of the BCPs only several inches from the tap point of origin and in the same panel), if the panel is not a listed assembly, then the NEC rules that apply are 240.21 (B)(1)(1), regarding conductor ampacity; 240.21 (B)(1)(2), pertaining to the limiting of the tap conductors to the device that the tap is meant to supply; and 240.21 (B)(1)(3), which states that raceway must be used from the tap to another enclosure . . . "Except at the point of connection to the feeder . . ."

Provided that this exception applies to the short runs that I have described, then maybe my angst over all this is much ado about nothing.

Best regards,
Shak

Actually the code is just a guideline until an AHJ officially adopts it as the rule, an even then they sometimes amend it.

 

[Shackled Designer]

That is an interesting and true point, kwired.

From the perspective of an OEM industrial panel designer, we are trying to better conform to the general code of the country to which we're sending our equipment and then hope that it is close enough to what the AHJ will require. Sometimes, a customer can be pretty specific on their local requirements, but often, they have no more idea of the particulars than we do.

In my memory, the toughest time given to us from an AHJ was on an install in Canada. My suspicion is that this may have had more to do with the particular inspector than with Canadian electrical codes, though I do have the impression that some Canadian codes are a bit more restrictive . . .

Thanks,
Shak

 

[kwired]

That is an interesting and true point, kwired.

From the perspective of an OEM industrial panel designer, we are trying to better conform to the general code of the country to which we're sending our equipment and then hope that it is close enough to what the AHJ will require. Sometimes, a customer can be pretty specific on their local requirements, but often, they have no more idea of the particulars than we do.

In my memory, the toughest time given to us from an AHJ was on an install in Canada. My suspicion is that this may have had more to do with the particular inspector than with Canadian electrical codes, though I do have the impression that some Canadian codes are a bit more restrictive . . .

Thanks,
Shak

My response to that is maybe you should get your product listed by a Canadian NRTL if you are selling it in Canada. I believe UL has a Canada division. Same for anything sold elsewhere. Install codes are more for people installing things, listings are for things that are already a complete unit - but may still need a supply source, interconnections for peripheral equipment, etc. that may have to be field installed/fabricated.