Would a slightly expanded (2020) 705.12(B)(3)(3) (sum of all breakers rule) be useful?

[wwhitney]

For the 2023 NEC I submitted PI 92-NFPA 70-2019 to expand 705.12(B)(3)(3) (sum of all breakers rule), so that the sum also excludes one breaker of the smallest size present. The physics on this is clear, it's still impossible to overload the bus under normal conditions.

This would allow, for example, a solar combiner box with, say, a 60A supply, a 60A busbar, (3) 20A OCPDs for microinverter circuits, and a 15A or 20A OCPD for monitoring equipment. It would also allow, say, a 100A panel with a 100A utility supply, a 100A breaker for on site generation (PV/ESS inverters), and a 100A breaker for a load panel (and no other breakers).

The CMP's response was "Adding additional features to this very simple rule creates unnecessary confusion for the enforcement and marking requirements." That's obviously a judgement call, and if the additional allowance would not be of practical use, I would agree. On the other hand, if it is something solar designers would regularly avail themselves of to simply installations and reduce materials used, I disagree.

So I'm asking practicing solar designers and others what they think about the idea, and if it would be useful to them, to submit a public comment on the issue by the closing date of August 19, 2021. Also please comment here so I can formulate my own public comment appropriately.

Thanks,
Wayne

 

[jaggedben]

There's not much payoff for the effort, is what I'd say. And call me cynical, but I think that 95% of electricians are already incapable of comprehending this label, and you would make it more complicated. I've gone as far as to slap that label over the unremoved twistouts of deadfronts, to drive home the point and increase the likelihood the label actually gets read. So the CMP has a point here.

The most meaningful change that I think could be realistic and worthwhile would be changing 120% to a larger number. Which would probably require presenting the CMP with empirical evidence of such safety, i.e. lab testing results. Perhaps also establishing a new or expanded UL standard for panelboards to allow such use or other specific uses. If we're gonna change the rules at this point, that's the direction I'd suggest.

 

[electrofelon]

The most meaningful change that I think could be realistic and worthwhile would be changing 120% to a larger number. Which would probably require presenting the CMP with empirical evidence of such safety, i.e. lab testing results. Perhaps also establishing a new or expanded UL standard for panelboards to allow such use or other specific uses. If we're gonna change the rules at this point, that's the direction I'd suggest.

I fully agree. That number is ridiculously overly conservative. IMO for a dwelling with a 200A service, we should be able to back feed up to the rating of the service entrance conductors. The CMP seems to be just stuck on theory and are clueless about actual dwelling loads and the way utilities have been supplying dwellings forever.

 

[Carultch]

For the 2023 NEC I submitted PI 92-NFPA 70-2019 to expand 705.12(B)(3)(3) (sum of all breakers rule), so that the sum also excludes one breaker of the smallest size present. The physics on this is clear, it's still impossible to overload the bus under normal conditions.

This would allow, for example, a solar combiner box with, say, a 60A supply, a 60A busbar, (3) 20A OCPDs for microinverter circuits, and a 15A or 20A OCPD for monitoring equipment. It would also allow, say, a 100A panel with a 100A utility supply, a 100A breaker for on site generation (PV/ESS inverters), and a 100A breaker for a load panel (and no other breakers).

The CMP's response was "Adding additional features to this very simple rule creates unnecessary confusion for the enforcement and marking requirements." That's obviously a judgement call, and if the additional allowance would not be of practical use, I would agree. On the other hand, if it is something solar designers would regularly avail themselves of to simply installations and reduce materials used, I disagree.

So I'm asking practicing solar designers and others what they think about the idea, and if it would be useful to them, to submit a public comment on the issue by the closing date of August 19, 2021. Also please comment here so I can formulate my own public comment appropriately.

Thanks,
Wayne

I agree with you that there should be provisions in this rule to allow for auxiliary circuits that are directly related to the system in a dedicated panelboard. Especially loads that are milliamps, and are insignificant to the loading of the busbar. Loads such as power monitoring & control systems, tracker controllers, surge suppression, AFCI/rapid shutdown systems.

There is also the issue that requiring you to sum up breakers is excessively conservative when it requires you to accumulate rounding errors. For instance, given 17 inverters that are 18A each, all on 25A breakers, adding up 25A instead of 1.25*18A will become significant. It causes a 400A panelboard to turn into a 600A panelboard, with no realistic scenario that will exceed 400A.

The rule that I'd recommend would be as follows:
Given a load center/panelboard/switchboard/switchgear that is dedicated to the PV system:
1. Busbar and main breaker sized to 125% of the sum of the inverter currents
2. Loads that are directly associated with the PV system, may also be added, provided that the total current of the loads adds up to less than 20% of the busbar rating.
3. The arrangement of breakers along the busbar shall not cause the current at any cross section of the busbar to exceed 80% of its rating
4. When the assembly is rated for continuous duty, 125% and 80% get replaced with 100%

 

[electrofelon]

The CMP's response was "Adding additional features to this very simple rule creates unnecessary confusion for the enforcement and marking requirements."

Just exempting circuit breakers used for monitoring and such doesn't seem to make things very complicated to me, what am I missing?

 

[jaggedben]

The sum of breakers rule as written has great benefits for applying to certain panels that are not dedicated to the PV system. So please don't propose to revise that rule to only apply to PV system panels. Add another rule if you want to make such a proposal.

 

[romex jockey]

IMO for a dwelling with a 200A service, we should be able to back feed up to the rating of the service entrance conductors.

dumb Q but compelled to ask ......does a backfeed register as added to an amp probe reading, or taking away from the total sum???

~RJ~

 

[Carultch]

dumb Q but compelled to ask ......does a backfeed register as added to an amp probe reading, or taking away from the total sum???

~RJ~

Subtraction. See Kirchhoff's current law.

Sources on opposite sides of a conductive material will subtract their current.
Sources on the same side of a conductive material will add their current.

 

[jaggedben]

dumb Q but compelled to ask ......does a backfeed register as added to an amp probe reading, or taking away from the total sum???

~RJ~

It will subtract from the load amps coming in, until they get to zero. When it exceeds the load amps, your amp reading will start going up again, the other way. However with just an amp clamp you won't be able to easily tell it's going in the other direction.

 

[wwhitney]

FWIW, the 2023 First Draft has changed 705.12(B) to start with "For power source connections to distribution equipment with no specific listing and instructions for combining multiple sources, one of the following methods shall be used to determine the required ampere ratings of busbars:"

This opens up the possibility of equipment being listed for multiple sources with greater latitude than provided by the current 705.12(B) rules. That of course raises the question of whether there is a listing standard for doing that. E.g. for a panelboard to be listed for two supplies from opposite ends of the busbar that jointly do not exceed, say, 150% of the busbar rating.

Cheers, Wayne

 

[pv_n00b]

For the 2023 NEC I submitted PI 92-NFPA 70-2019 to expand 705.12(B)(3)(3) (sum of all breakers rule), so that the sum also excludes one breaker of the smallest size present. The physics on this is clear, it's still impossible to overload the bus under normal conditions.

This would allow, for example, a solar combiner box with, say, a 60A supply, a 60A busbar, (3) 20A OCPDs for microinverter circuits, and a 15A or 20A OCPD for monitoring equipment. It would also allow, say, a 100A panel with a 100A utility supply, a 100A breaker for on site generation (PV/ESS inverters), and a 100A breaker for a load panel (and no other breakers).

What if some electrician tries to get cute and puts four 20A OCPDs for inverters in the 60A panel, then just subtracts the smallest CB, a 20A? The devil is in how it can be misused.

 

[romex jockey]

It will subtract from the load amps coming in, until they get to zero. When it exceeds the load amps, your amp reading will start going up again, the other way. However with just an amp clamp you won't be able to easily tell it's going in the other direction.

thx Jag,
I'm really starting to think i should have paid more attention to the electron theory part of my schooling......

~RJ~

 

[wwhitney]

What if some electrician tries to get cute and puts four 20A OCPDs for inverters in the 60A panel, then just subtracts the smallest CB, a 20A? The devil is in how it can be misused.

If the sum of the inverter currents on the (4) 20A OCPDs exceeds 48A, that would be a violation of 690.8.

But as far as protecting the busbar from overload, (4) 20A inverter OCPDs is not a problem. The 60A breaker protects the 60A busbar.

With (4) 20A breakers and a 60A breaker on a busbar, there's no way to get a busbar current in excess of 60A when each breaker is carrying a current (in either direction) not exceeding its rating.

Cheers, Wayne

 

[pv_n00b]

But you see how the proposed language is not clear about what is intended. What is intended is that a smaller load be allowed in these panels that would otherwise not be allowed.

 

[wwhitney]

But you see how the proposed language is not clear about what is intended. What is intended is that a smaller load be allowed in these panels that would otherwise not be allowed.

Sorry, I don't see, perhaps you can explain. The proposed language is not in this thread, it's in the PI. And rather than "smaller" it's "not larger than any other"; and it doesn't have to be a load, the rule doesn't differentiate between loads and supplies.

The algorithm I propsose is (for the simple case of only 2 pole breakers in a single phase panel): take all the breaker ratings. Throw away the rating of the main breaker protecting bus. Sort them in descending order; throw away the last entry on the list (that's the proposed change). Now add them up and compare to the bus rating.

So for the PI I proposed this change in language:

The sum of the ampere ratings of all overcurrent devices on panelboards, both load and supply devices, excluding the rating of the overcurrent device protecting the busbar, and excluding one overcurrent device of the smallest size present, shall not exceed the ampacity of the busbar.

Is that not clear? If there's a way to express it more simply, I'm very interested.

Cheers, Wayne

 

[pv_n00b]

I agree with the CMP, there is a simplicity to (3)(3) now and that's not something we can same about much of 690 or 705. I also agree that the system you propose would not overload the busbar. But being factual will not always win over the CMP.

 

[jaggedben]

...
Is that not clear? If there's a way to express it more simply, I'm very interested.

...

It's as clear as can be, to people in possession of a diploma from a four year college.

 

[ggunn]

The sum of the ampere ratings of all overcurrent devices on panelboards, both load and supply devices, excluding the rating of the overcurrent device protecting the busbar, and excluding one overcurrent device of the smallest size present, shall not exceed the ampacity of the busbar.

It seems imprecise to me. The smallest breaker could be 20A or it could be 100A on the same busbar with everything else the same. A better way might be to increase the 100% number by a fixed amount.

 

[wwhitney]

It seems imprecise to me. The smallest breaker could be 20A or it could be 100A on the same busbar with everything else the same.

Yes, that's intentional, as that is what the combinatorics tells us. This is for the model where each breaker may carry current up to but not exceeding its rating, and each breaker might be either a supply or a load (all possibilities need to be considered). Then if you want to ensure the busbar current at any cross section never exceeds the busbar rating, while allowing maximum flexibility, It really needs to be "one breaker of the smallest size present," no more and no less.

Example: 100A bus, 100A main breaker, (2) 40A breakers, (1) 20A breaker. This is allowed under the present rule. Consider a two column tally of supply breakers and load breakers. Worst case would be the 100A breaker on side, and 40A + 40A + 20A on the other side; each column tallies to 100A, the busbar rating.

But current is neither sourced nor sinked to the busbar, so for the actual currents flowing through each breaker, the corresponding sums must equal. That means when the two columns of breaker ratings have unequal totals, the actual current in the busbar is at most the lesser of the two column sums. [That's why I divided the breakers evenly for the worst case above.] Therefore it is enough to ensure that one of the column totals doesn't exceed the busbar rating.

So add a 20A breaker to the above example. You can put it in one column, causing that column's total to exceed 100A. But the other column will still be only 100A, and the busbar can only see 100A. There's no way to increase both column totals with just one more breaker of the smallest size present.

Now instead add a 30A breaker to the initial example. I can rearrange the columns to be 40A + 40A + 30A vs 100A + 20A. The lesser sum is 110A, which exceeds the busbar rating. No good.

Cheers, Wayne