As will be clear in the 1999 NEC, a power panelboard
is any panelboard that is not a lighting and appliance
branch-circuit panelboard. Exactly where the Code draws that
line has not been clear at all, however, which is why CMP- 9
has been reworking the definitions. The distinction has two
practical effects. First, a lighting and appliance
branch-circuit panelboard needs individual protection for its
busbars, whereas other panelboards can be protected simply by
assurance of appropriate loading through an Article 220-load
calculation. Secondly, lighting and appliance branch-circuit
panelboards carry a 42-circuit limitation.
The problem with the 1996 NEC
The 1996 NEC definition, in Section 384-14, reads as
follows: "For the purposes of this article, a lighting
and appliance branch-circuit panelboard is one having more
than 10 percent of its overcurrent devices rated, 30 amperes
or less, for which neutral connections devices are
provided."
The question before the panel was over the meaning of the
last clause of that sentence. Did it mean that the neutral
connections were provided for the panelboard (i.e., a
terminal bar with a connection to the feeder neutral)? On the
contrary, did it mean that neutral connections were utilized
by the branch circuit that extended from the 30-ampere or
lower overcurrent devices? For example, suppose you wire a
240-volt heat pump (no neutral used in the circuit) on a
30-ampere branch circuit. Does that overcurrent device qualify
as one that would be counted toward the 10 percent threshold,
assuming there are vacant neutral connections otherwise
provided within the panelboard for other possible connections?
It turns out that there are definitely two very large
schools of thought on this issue, and we discovered that they
hadn’t been talking to each other. Manufacturers have to
meet special product standard requirements in making suitable
panels for lighting and appliance use. They routinely apply
the second interpretation, which is entirely dependent on
field wiring connections. They couldn’t imagine anyone
saying the mere presence of unused neutral terminals might
qualify the panel as a lighting and appliance panel.
On the other hand, many others applied the first
interpretation, since in terms of equipment, neutral
connections cannot be provided for most overcurrent devices
(excepting GFCI or GFPE or switching-neutral breakers), only
for the panelboard in which the overcurrent device is
installed. Remember, the present literal text does not use the
term "circuits" and therefore the antecedent of the
pronoun "which" mere logically applies to the
equipment mentioned in the first part of the definition.
In the 1996 code making cycle, CMP-9 made a stab at this,
accepting clarifying language in the direction of the second
interpretation during the proposal period. The manufacturers,
citing the realities of the product standards, strenuously
objected during the comment period. The panel decided to hold
the material and study it for the 1999 cycle. We came up with
a solution for the 1999 NEC, that we hope will please most of
the users.
The Safety issues
Over the next three years, l served on a CMP-9 task group
charged with considering the following issues, which we put
before the industry in the ROC for the 1996 NEC:
1. Should the Code continue to maintain a special class of
panelboards? The "lighting and appliance
branch-circuit" designation is, at present, a vehicle for
providing enhanced protection in the form of individual
protection and circuit limitation for some panelboards. Should
these rules simply apply to all panelboards? Is there a solid
technical basis for saying that certain types of installations
do not need the additional restrictions?
2. If there is a basis for a separate classification, how
could that code be written to squarely address those concerns
and, to the extent possible, avoid the following anomalies in
the existing test?
a. The present rules don’t include a delta-connected
panelboard (no neutral) in the definition, even if the entire
load consists of lighting and appliances on 30-ampere or
smaller circuits. In addition, the actual definition does not
reference branch circuits. This makes even the term
"lighting and appliance branch-circuit" misleading
and inaccurate.
b. The present rules depend on the existence of neutral
connections; to what extent do these connections actually
affect relative safety? The task group understands that
neutral connections facilitate single-pole connections that
may load panel buses unequally, but so could two-pole
connections in a three-phase panel.
c. The present rules are inconsistent with feeder
overcurrent device calculations for motor loads per Sections
430-62(a) and 430-63, particularly where a large motor is
connected to a panelboard that falls within the lighting and
appliance definition.
3. Whether the rules expand to cover more panelboards or
not, the specific restrictions also need to be reviewed:
a. Individual protection. The task group does understand
the potentially serious implications of expanding the
individual protection rule, particularly where main-lug-only
panelboards are now connected to ten foot taps. In these cases
the only existing protection is an accurate load calculation,
but is there loss experience to justify expanding the
individual protection requirements for even more panelboards?
Conversely, the task group agrees that service applications
need not be further restricted because they are subject to
special wiring methods and location restrictions in Article
230, and unqualified persons are less likely to tamper with
them.
b. Maximum circuit limitations. Are 42 circuits a
defensible limitation with existing technology? The limitation
originated in the 1933 NECâ, and was apparently based on a
desire to limit the amount of combustible wiring associated
with any one panelboard. Based on the limited material
unearthed by the task group to date, this rule was apparently
motivated by a fire in a New York City hotel back when the
only available conductor insulation was combustible rubber.
Clearly this is no longer appropriate substantiation to
maintain the rule, however, perhaps there are other concerns,
such as mutual heating effects of large numbers of conductors
in a confined space. The panel previously rejected an attempt
to remove this limitation in the 1971 cycle, but only with the
assertion that it felt that it is necessary for safety.
4. In summary, the most credible argument in support of
maintaining different classes of panelboards will probably
turn on a judgment that certain applications are significantly
more likely to be subject to improperly calculated and
inspected circuit additions. If this conclusion was reached
(and it may well not be), how should the Codeâ take this into
account? In practice, should the restrictions apply to certain
types of panelboards? Should the distinction turn instead, for
example, on the type of occupancy involved or on some other
criteria? The task group is aware of an increasing reluctance
in some code making circles to sanction yet another example of
a separate rule for industrial occupancies (in industrial
occupancies, where conditions of maintenance and supervision
assure that only qualified individuals will service . . .
etc.)
These questions have been asked before. For example, the
present language in Section 384-16 (a) (requiring individual
protection) essentially dates from the 1965 Codeâ, and the
substantiation in the "Preprint" at the time says:
"The NI Fact Finding Report (UL File NC492) shows that
overheating of copper parts and insulated conductors can
result if loading of a panelboard is increased beyond the
nameplate rating by indiscriminate adding of branch circuit
loads. Main overcurrent devices for the panel will prevent
overheating of the panel."
"CMP- 9 wrestled with expanding this rule to cover all
panelboards. Although a slight minority favored the idea at
the time, there was no consensus. One negative voter (Pringle,
NEMA) said that ‘Power panelboards usually have overcurrent
devices of sufficiently high rating that overheating is not a
problem.’ The concept was held for further study, and then
unanimously rejected for want of adequate substantiation in
the 1968 cycle."
The 1999 NEC result
First and most importantly, CMP- 9 decided to retain a rule
that mandates enhanced protection for certain panelboards. The
panel agreed that panelboards with appreciable numbers of
small branch circuits and neutral connection present (whether
or not utilized) appear to be the ones that are most likely to
have additional circuits and loads added from time to time and
under the least rigorous supervision or likelihood of
inspections. However, the panel also did not want to upset the
present product standards that only use the "lighting and
appliance" terminology when the neutrals that are counted
are actually part of the branch circuits.
Our solution was to clarify (the definition in the way the
manufacturers are using the term neutrals only counted where
part of the smaller branch circuits) and then to include the
other panelboards with small branch circuits (30 ampere or
less; neutrals considered even if not used at the time). We
needed a "handle" for these other panelboards,
rather than using the awkward "other than a lighting and
appliance branch-circuit panelboard." We picked the term
"power panel" which is in common use for this
purpose already. It was a simple manner to define it in a way
that is exactly complimentary to the lighting and appliance
branch-circuit panelboard definition. The 1999 NECâ version
of Section 384-14 will read as follows:
"384 14. Classification of Panelboards. Panelboards
shall be classified for the purpose of this article as either
lighting and appliance branch-circuit panelboards or power
panelboards.
"(a) Lighting and Appliance Branch-Circuit Panelboard. A lighting and appliance branch-circuit panelboard is one
having more than 10 percent of its overcurrent devices
protecting lighting and appliance branch circuits. A lighting
and appliance branch circuit is a branch circuit that has a
connection to the neutral of the panelboard and that has
overcurrent protection of 30 amperes or less in one or more
conductors.
"(b) Power Panelboard. A power panelboard is one
having 10 percent or fewer of its overcurrent devices
protecting lighting and appliance branch circuits."
To cover the power panelboards with small branch circuits,
we modified Section 384-16 by adding a new subsection (b). We
decided that the small delta-connected panels could be
included with their larger brethren, since that kind of
distribution is usually restricted to more sophisticated
industrial occupancies. The new subsection is as follows:
"(c) Power Panelboard Protection. In addition to
the requirements of Section 384-13, a power panelboard with
neutral connections available and more than 10 percent of its
overcurrent devices protecting branch circuits rated 30
amperes or less shall be protected on the supply side an
overcurrent protective device having a rating not greater than
that of the panelboard.
"Exception: This individual protection shall not be
required for a power panelboard used as service equipment with
multiple disconnecting means in accordance with Section
230-71. "
The result is similar requirements on individual protection
for both power panelboards and lighting and appliance
branch-circuit panelboards in cases where uninspected future
alterations seem likely: Where more than 10 percent of the
overcurrent devices don’t exceed 30 amperes, and where
neutral connections are easily available for further use,
individual protection is required. Note that we didn’t
change Section 385-15. This means that although you’ll need
individual protection for the power panels with smaller
circuits, you won’t need to observe the 42-circuit
limitation. After requiring individual protection, we didn’t
think there was adequate substantiation to impose the further
limitation.
We based the exception on a judgment that if one of these
small panels were part of the service equipment, any
modification would be likely to have much better supervision
and inspection that the same equipment used as a remote
subpanel. The last issue we considered was whether the
allowance in Section 430-62(a) and 430-63 for increased sizing
on feeder protective devices should be carried over to this
part of the Code. We decided that since we are clarifying the
restrictions on certain panels, because we think they may be
abused and go uninspected, it would be inconsistent to then
turn around and allow the same panels to have their
overcurrent protection calculated under a section of the NEC
that requires a very sophisticated analysis and inspection to
be sure that it is properly applied. You can still use the
provisions in Article 430 for large power panels without the
smaller branch circuits.
This is a conservative approach, with minimal effect on
product standards while adequately addressing the safety
concerns. We felt that more radical steps, such as eliminating
the "lighting and appliance branch-circuit
panelboard" altogether, should await more substantiation
that is adequate. Interestingly, we received no public
comments that provided any such substantiation.
Anthony Montuori has been a member of both the Master
Electricians Licensing Board and New York City Advisory Board
since 1984. He was the New York Chapter chairman of IAEI, and
a NFPA member. Tony is a past Kiwanis club president, and is
on the Blue Ribbon Advisory Council of John Jay High School
|
