#14 Cu on a 100A breaker

[wwhitney]

(a) 430.24 says the feeder conductors must have an ampacity of at least 200A.
(b) 430.62 has options A and B
(A) would say that the feeder OCPD must be at most 100A.
430.62(B) Can apply where feeder conductors have an
ampacity greater than required by 430.24.
We'll choose 4/0 Cu with 90C insulation, 75C ampacity 230A.

(c) 430.63 says that the feeder OCPD must be at least 250A.
QED

So your result is that the feeder conductor size determined by 430.24, 3/0 Cu with 75C ampacity 200A, is actually not allowed, and you have to bump up one size to 4/0 Cu?

That's way less plausible than the common understanding that the word "only" is implicit in 430.62(A), as in "a feeder supplying only a specific fixed motor load(s)." I just made a PI special for you, it changes 430.62(A) to "a feeder supplying only fixed motor load(s)" to improve clarity.

I would agree with your understanding of the wording in 430.62 if 430.63 were absent from the NEC. Except that then 430.62 would need to be fixed to add in the non-motor loads.

Cheers, Wayne

 

[tortuga]

I just made a PI special for you, it changes 430.62(A) to "a feeder supplying only fixed motor load(s)" to improve clarity.

Yes great idea thanks for doing that, I enjoy your thought experiments. I think that deserves another milkshake or beverage of your choice.
Next time I get to Berkley

 

[brycenesbitt]

In my area we still have some Edison base fuse boxes feeding dwelling loads

And we have test results showing how robust and reliable that fuse based system is, was, and will continue (unless someone hacks it up and messes with it).

 

[ramsy]

And we have test results showing how robust and reliable that fuse based system is, was, and will continue (unless someone hacks it up and messes with it).

Much more reliable than AFCI, or GFCI. Fuses work with harmonic loads, without constant nuisance trips, or reset-button failures.

 

[jim dungar]

Much more reliable than AFCI, or GFCI. Fuses work with harmonic loads, without constant nuisance trips, or reset-button failures.

You are not comparing similar items.
Fuses, like thermal magnetic breakers, only provide over current and short circuit protective functions.
If you had fuses that were sensitive down to 6mA, I would expect similar nuisance tripping as we find with GFCI breakers.
AFCIs are a different animal entirely.

 

[jim dungar]

Actually the extra items clarify the first item, that is the purpose of an adjectival phrase.

For my own edification, I went to AI and diagramed 430.62(A).
I forgot about participial phrases, but I had the modifing subject part correct.

This is an area where AI could help when creating NEC articles.

Sentence diagram (structured breakdown)​

Main clause

• Subject (implied): [A feeder]
• Verb (predicate): shall be provided
• Direct object / complement: with a protective device having a rating or setting not greater than ...

Modifiers of subject

• "supplying a specific fixed motor load(s)" — participial phrase modifying "feeder"
• "and consisting of conductor sizes in accordance with 430.24" — participial phrase modifying "feeder" (coordinated with the previous modifier)

Main complement (what the feeder is provided with)

• "a protective device" — head noun of the complement
  • Postmodifier: "having a rating or setting not greater than the largest rating or setting of the branch-circuit short-circuit and ground-fault protective device for any motor supplied by the feeder, plus the sum of the full-load currents of the other motors of the group."

 

[don_resqcapt19]

For my own edification, I went to AI and diagramed 430.62(A).

I have said many times that the code language should be provided in the diagrammed form to the CMPs before the final vote.

 

[Elect117]

I tried to get myself up to speed with the discussion but forgive me if I missed anything or misread something.

This is for ground fault and short circuit protection of feeder motor circuit conductors. This does not provide overload protection for the loads or the conductors. That is required to be considered elsewhere. I agree that the wording of "shall have a rating not less than" is actually more of an issue. You should be able to select a device that can do both without needing to oversize it just to meet 430.63. I think the idea was that motors should be able to start up safely with other loads sharing the circuit's feeder.

But I don't know if there is a situation that could have feeder overload protection/feeder conductor protection and meet 430.63 without needing to upsize the conductors to compensate. So if your minimum per 430.63 is 200A, then you would more than likely be stuck with 3/0 and your taps per 240.21(B) + 430.62/430.52 and your other load to 240.21(B) + any of their respective sections in the code.

"
430.63 Rating or Setting — Motor Load and Other Load(s).

Where a feeder supplies a motor load and other load(s), the feeder protective device shall have a rating not less than that required for the sum of the other load(s) plus the following:

• (1) For a single motor, the rating permitted by 430.52
• (2) For a single hermetic refrigerant motor-compressor, the rating permitted by 440.22
• (3) For two or more motors, the rating permitted by 430.62

Exception:
Where the feeder overcurrent device provides the overcurrent protection for a motor control center, the provisions of 430.94 shall apply.

"

 

[suemarkp]

I have said many times that the code language should be provided in the diagrammed form to the CMPs before the final vote.

Or if it was written like computer code it would be easier and definitive using if-then-else, lists, and-or-not, and parenthesis to group modifiers. The compiler provides only one answer with computer code, whether what was written was your intention or not.

 

[wwhitney]

But I don't know if there is a situation that could have feeder overload protection/feeder conductor protection and meet 430.63 without needing to upsize the conductors to compensate.

Agreed with 430.63 as written. But this is just evidence that 430.63 is wrong.

What we reasonably expect to happen with the NEC requirements (for a feeder, where the non-motor loads will have downstream branch circuit protection that provides overload protection for the load, and with inverse time breakers for OCPD):

40A FLC motor load alone: minimum conductor ampacity is 50A, maximum OCPD is 100A (ignoring 430.52(C) exceptions)

As above plus 25A non-continuous non-motor load: both of these values increase by 25A, so minimum conductor ampacity is 75A, maximum OCPD is 125A.

Basically at a big picture level and not being too exact with the details, the way I think about it is that every conductor gets an allowable SC/GF to Overload discrepancy. That's a "breaker oversizing allowance", the difference between the allowable SC/GF breaker rating and the actual conductor ampacity. That factor is one allowance of the load (125% of the FLC) of the largest motor supplied by the conductor. That's from the difference between the 250% factor in Table 430.52 (for most of the motor types) and the 125% factor for ampacity calculation.

Cheers, Wayne

 

[Elect117]

I don't know if the load allows for an "overload discrepancy" as much as it is trying to keep the OCPD device from tripping on motor start up. I don't know if they equate.

But the feeder conductors and their protective device for overload is vague in 430. I don't know if it should be addressed in like 215 or what. But in my opinion the only mention of feeder SC/GF and overcurrent selection should be something like this:

Motor Feeder SC/GF device selection should coordinate with the selected branch circuit SC/GF device and be able to sustain the start up conditions of motors and nonmotor loads on the feeder.

 

[wwhitney]

I don't know if the load allows for an "overload discrepancy" as much as it is trying to keep the OCPD device from tripping on motor start up. I don't know if they equate.

When you upsize a breaker to avoid motor startup tripping, you are changing the OCPD size for SC/GF to something bigger than the OCPD size required for overload. That difference is what I called the "SC/GF to Overload" discrepancy. That breaker no longer provides the necessary overload protection for the downstream conductors, but that's OK, as overload protection can be provided by downstream devices, unlike SC/GF protection.

But in my opinion the only mention of feeder SC/GF and overcurrent selection should be something like this:

Motor Feeder SC/GF device selection should coordinate with the selected branch circuit SC/GF device and be able to sustain the start up conditions of motors and nonmotor loads on the feeder.

That idea would work great if our electrical system had two different sets of devices, one that provides overload protection and another that provides SC/GF. Then we could have separate rules for the two different sets of devices as to sizing and location in the circuit.

But we don't have that, we just have a single class of overcurrent devices that usually provide both SC/GF and overload protection. This is less flexible, and in the situations like motors where that lack of flexibility is a problem, the NEC needs to provides rules for when and how the overcurrent devices are only required to provide SC/GF and thus may be oversized relative to ampacity.

Cheers, Wayne

 

[Elect117]

When you upsize a breaker to avoid motor startup tripping, you are changing the OCPD size for SC/GF to something bigger than the OCPD size required for overload. That difference is what I called the "SC/GF to Overload" discrepancy. That breaker no longer provides the necessary overload protection for the downstream conductors, but that's OK, as overload protection can be provided by downstream devices, unlike SC/GF protection.



That idea would work great if our electrical system had two different sets of devices, one that provides overload protection and another that provides SC/GF. Then we could have separate rules for the two different sets of devices as to sizing and location in the circuit.

But we don't have that, we just have a single class of overcurrent devices that usually provide both SC/GF and overload protection. This is less flexible, and in the situations like motors where that lack of flexibility is a problem, the NEC needs to provides rules for when and how the overcurrent devices are only required to provide SC/GF and thus may be oversized relative to ampacity.

Cheers, Wayne

Maybe a PI to add 430.63 to 430.62 in some fashion. They have been going for the "consolidated list" stuff as of late. That might solve the open ended issue.

I really do think the intent is to have a conductor sized to 430.24 and if the device is doing both SC/GF and overload then have the conductor size increased to the overload for feeders. Depending on how the feeder is run, I guess. The whole thing with the MCC and their lack of mention of anything else just throws me off. The MCC needs to oversize the bus based on the GF/SC and related overload protection but a switchgear wouldn't? Or panelboard wouldn't? Or feeder wire?

I also think you could put overload protection at the point of supply for the branch circuit and it would adequately protect the loads if the wire was size to 430.24. But I don't know if it is addressed clearly. They considered the sizing of feeder conductors and the SC/GF protection but ignore the overload protection part and just don't mention it. The only mention of overload is in branch circuits with motors.

 

[wwhitney]

I really do think the intent is to have a conductor sized to 430.24 and if the device is doing both SC/GF and overload then have the conductor size increased to the overload for feeders.

To clarify, are you saying that you think the intent for feeders (430 Part V) is:

40A FLC motor only: OCPD at least 50A. For 50A - 100A OCPD, 50A ampacity suffices. For larger OCPD, 240.4 applies.
40A FLC motor plus 50A non-continuous non-motor: OCPD at least 100A. Ampacity at least 100A. Must comply with 240.4.

I don't, I think for the last one it should be:

OCPD at least 100A. For 100A - 150A OCPD, 100A ampacity suffices. For larger OCPD, 240.4 applies.

Here's why that's OK and in accordance with the other motor allowances in the NEC. Simplest model is a feeder to a panel with a breaker for the motor and typically a 50A breaker for the non-motor load. The motor breaker can be 50A to 100A, and the motor branch circuit is only required to have an ampacity of 50A. The motor overloads protect the motor branch circuit from overload. So the branch circuit conductors have proper overload protection at 50A, and proper SC/GF protection at 50A-100A.

That means for the feeder, overload protection is provided at 100A, in the form of 50A from the motor overloads and 50A from the non-motor load breaker. In which case the breaker at the source of the feeder just needs to provide SC/GF protection, and 100A-150A is sufficient for that.

Now admittedly the non-motor load could be on, say (4) 20A branch circuits. So the overload protection is at 130A which seems too high. But there are other places in the NEC where overload protection is dependent on load calculation, such as 230.90(A) Exception (3). So it's a judgement call as to whether this phenomenon of insufficient overload protection like this is common enough and problematic enough to not rely on overload protection via load calculation.

Cheers, Wayne

P.S. The cleanest way to handle all this, and maybe feeder taps too, would be to have separate rules for each point of the wiring system on the required SC/GF protection (which must be provided upstream) and overload protection (which may be provided downstream). Then you'd comply with both.

Or an even simpler way would be if we had OCPD whose behavior over all time periods better matched the conductor damage curve shape, so that a 40A FLC motor wouldn't trip a 50A OCPD at startup. Then there would no need to consider overload protection and SC/GF protection separately.

 

[Elect117]

To clarify, are you saying that you think the intent for feeders (430 Part V) is:

40A FLC motor only: OCPD at least 50A. For 50A - 100A OCPD, 50A ampacity suffices. For larger OCPD, 240.4 applies.
40A FLC motor plus 50A non-continuous non-motor: OCPD at least 100A. Ampacity at least 100A. Must comply with 240.4.

I don't, I think for the last one it should be:

OCPD at least 100A. For 100A - 150A OCPD, 100A ampacity suffices. For larger OCPD, 240.4 applies.

I don't think there is anything wrong with either. I don't think the code is clear and because of that, you might run into a situation where both work or only one or the other works and we might need to air on the side of protecting the feeder conductors at their rating.

For example, a 400A 120/208V 3ph panel board with a mixture of motors, AC equipment for freezers, freezer door heaters and maybe some forklift chargers. The largest motor is 150A (with 125%) and the other loads (continuous or otherwise) added up to 300A.

IF you used 430.62(A) as the method (2.5* largest motor + everything else) you could end up over 400A, exceeding the panel. So you would either, in my opinion, need to:

1) increase the panel size
2) install conductors and ocpd to 400A to protect panel (if it did not trip on motor start up).

I wouldn't say you could 2x1/0 Cu or 2x2/0 and a 400A breaker for that panel. 400A OCPD meets 408.36 but the terminations at the load end could fail prior to that protecting the feeder. So if they plugged in some higher VA chargers than the plans originally marked then we are getting to where the overload protection could have saved property/equipment if it was not oversized to the wire for the one motor.

But I am on the fence. I know in some circumstances there would be no likelihood of overloading the wire. But in others, where maybe the motors get larger and the other loads are 50A receptacles, there could end up being issues. Primarily where the loads can be changed without needing to call someone.

 

[tortuga]

I don't know if the load allows for an "overload discrepancy" as much as it is trying to keep the OCPD device from tripping on motor start up. I don't know if they equate.

But the feeder conductors and their protective device for overload is vague in 430. I don't know if it should be addressed in like 215 or what. But in my opinion the only mention of feeder SC/GF and overcurrent selection should be something like this:

Motor Feeder SC/GF device selection should coordinate with the selected branch circuit SC/GF device and be able to sustain the start up conditions of motors and nonmotor loads on the feeder.

Fundamentally and from a physics perspective, I *think* we can all agree there are three main types of circuit protection: short-circuit, ground-fault (instantaneous or magnetic types), and overload (thermal type) protection. Article 430 splits those functions into two separate devices for motor branch circuits (or allows for a split). My take on the intent is its not to have 'split protection' on a feeder itself because there is no 'Feeder Overload Relay.'

My answer in post 59 is based on reviewing the 1950 and 1953 NEC changes where these sections on motor feeder protection were first added, since motor branch circuits may have split protection (overload protection at the motor controller and an OCPD sized for only ground-fault and short-circuit protection), the minimum feeder size and OCPD must match the largest allowed motor OCPD, including the conductor size. If other loads are powered from this feeder, the Article 220 calculated load may be added to this size.
My take is in the early 50's the committees created what we now call 430.62 and 430.63 to accomplish two goals for feeders: to establish a 'Maximum' or 'Ceiling' to ensure that, while we allow 'room to start,' we do not leave the 'door open' for a fire, and to supply a feeder large enough for the largest motor to start. There does not appear to be a rule allowing a feeder to be sized the same way as a motor branch circuit, as a feeder OCPD serves as its own overload protection.

Consequently, the more one upsizes the largest motor branch circuit OCPD, the larger the feeder conductors must be. In modern applications involving VFDs and soft starts, this is often a non-issue; most feeders are calculated at 100% of the non-continuous load plus 125% of the continuous load plus 25% of the largest motor or something like that. However, if an older motor leverages a significantly large branch breaker, it may, by my interpretation force the upsizing of the entire feeder.

 

[wwhitney]

I don't think the code is clear

Agreed, but it needs to be. It should give you a deterministic procedure so that you can input the loads and type of loads on a feeder, and it outputs the set of allowable OCPD sizes, and the minimum conductor ampacity for each of those sizes. If necessary, that procedure can depend on the arrangement of downstream OCPD sizes (to determine what level of overload protection that arrangement provides to the feeder).

For example, a 400A 120/208V 3ph panel board with a mixture of motors, AC equipment for freezers, freezer door heaters and maybe some forklift chargers. The largest motor is 150A (with 125%) and the other loads (continuous or otherwise) added up to 300A.

If those are all 3 phase loads, and Article 220 spits out a total load of 150A + 300A = 450A, then the panelboard is too small, so I'm not clear on why this would be an interesting example. Did you mean that the other loads are only 150A, not 300A?

Cheers, Wayne

 

[wwhitney]

My take on the intent is its not to have 'split protection' on a feeder itself because there is no 'Feeder Overload Relay.'

Certainly for a feeder that powers motors only, 430.62(A) allows split protection. So perhaps you mean that for a feeder that powers mixed loads?

I find it an unreasonable result that if you have a feeder supplying only a 90A FLC motor, you could use 113A ampacity conductors protected at 225A, but that if you add a 5A non-continuous non-motor load to that, you now need 226A ampacity conductors protected at 250A. Adding 5A of load should not cause a 113A increase in minimum ampacity size.

There does not appear to be a rule allowing a feeder to be sized the same way as a motor branch circuit, as a feeder OCPD serves as its own overload protection.

It need not, the arrangement of downstream OCPD could provide overload protection.

Cheers, Wayne

 

[Elect117]

If those are all 3 phase loads, and Article 220 spits out a total load of 150A + 300A = 450A, then the panelboard is too small, so I'm not clear on why this would be an interesting example. Did you mean that the other loads are only 150A, not 300A?

Cheers, Wayne

I meant 300 total. 150A motor with 150A of various other loads.

 

[tortuga]

Certainly for a feeder that powers motors only, 430.62(A) allows split protection.

As I explained thats not the way I interpret it, but it will be interesting to see what happens with your PI, I think we all agree it could be more clearly addressed.