2014 National Electrical Code Changes

As we have done for quite a number of years now, we’ve teamed up with our long-standing NEC Consultant and Code guru Mike Holt to present you with a summary of the key changes made during this revision cycle. Although this article is nearly 25 pages long, the time you spend reading it will be well worth it. Not only will it save you from reviewing all 900+ pages of the actual NEC, but it will also help you gain an immediate understanding of the changes that affect the largest number of our readers.

This group of 20 key revisions focuses on the topics of selective coordination, field markings, dedicated spaces, multiwire branch circuits, GFCI and AFCI protection, conductor sizing, lighting load calculations, building disconnects, grounding electrodes, conductor ampacity, raceway supports, switch connections, grounding and bonding transformers, receptacles and plugs in hazardous locations, wiring methods in special occupancies, disconnects for sign lighting, wiring for spas and hot tubs, and emergency illumination requirements.

As you read through the analysis, please note that the underlined text is new to the Code. Although it might be slightly reworded or shortened from the actual text in the NEC, it’s a good representation of the intent of the real rule change.

Change No. 1: Coordination (Selective)

A revised definition and clarification of the type of person who can design a selectively coordinated system will address arguments that have been made on this topic.

Article 100 Definitions
Part I. General
Coordination (Selective). Localization of an overcurrent condition to restrict outages to the circuit or equipment affected, accomplished by the choice of overcurrent devices. Selective coordination includes all currents from overloads to short circuits. (Fig. 1).

Author’s comments: Selective coordination means the overcurrent protection scheme confines the interruption to a particular area rather than to the whole system. For example, if a short circuit or ground fault occurs with selective coordination, the only breaker/fuse that will open is the one protecting just the branch circuit involved. Without selective coordination, an entire floor of a building can go dark.

Analysis: Ever since its introduction into the Code, few rules have created more arguments than selective coordination. Selectively coordinating a system isn’t a walk in the park if circuit breakers are involved, but it can be done. In order to allow a wider range of products to be in a given installation, many jurisdictions have created legal definitions of the term by providing a numerical value to the time involved to clear a fault. For example, “systems shall be selectively coordinated to within XX seconds.” This has never been the intent of the NEC. Reading the dozens and dozens of rejected proposals to 700.27 and 701.27 in this change cycle confirms this fact. The change to the definition doesn’t technically change anything, but it may help people who aren’t aware of the fact that selective coordination is intended to apply to the full gamut of currents — not just overloads but ground faults and short circuits too.

The Code only requires selective coordination for the most important systems, such as emergency and legally required standby systems and health care facilities (see 517.17, 700.27, and 701.27). Due to the importance of these systems, the electrical installation must be a complete 100% coordinated system. There’s no time limit or allowance for short overlaps of the overcurrent devices. A selectively coordinated system must ensure no possibility whatsoever of an upstream overcurrent device, such as the main breaker for the building, opening before the overcurrent device closest to the fault opens.

In addition to the definition change, the selective coordination of an emergency system must now be designed by an engineer or similarly qualified person.

Article 700 Emergency Systems
Part VI. Overcurrent Protection
700.28 Selective Coordination. Overcurrent devices for emergency power systems must be selectively coordinated with all supply-side overcurrent devices. The design must be made by an engineer or similarly qualified person, and it must be documented and made available to the AHJ.

This same design requirement has also been added for legally required standby systems.

Article 701 Legally Required Standby Systems
Part IV. Overcurrent Protection
701.27 Selective Coordination. Overcurrent devices for legally required standby systems must be selectively coordinated with all supply-side overcurrent devices. The design must be made by an engineer or similarly qualified person, and it must be documented and made available to the AHJ.

Analysis: Since its inception in 2005, few Code rules have had as many proposals as this one (for both emergency and legally required standby systems). Very few of these proposals pass because most of them seek to reduce the safety requirements. This change, however, makes the requirement a bit more stringent, as it now requires the design of the system to be engineered by someone who knows what they’re doing. This was probably already occurring, but now an inspector or plans examiner has a rule they can hang their hat on if the design is beyond the competency of the designer.

Change No. 2: Markings

New requirements for field-applied labels have been added to 110.21.

Article 110 Requirements for Electrical Installations
Part I. General
110.21 Markings
(A) Manufacturer’s Markings. The manufacturer’s name, trademark, or other descriptive marking must be placed on all electrical equipment and, where required by the Code, markings such as voltage, current, wattage, or other ratings must be provided. All marking must have sufficient durability to withstand the environment involved.
(B) Field-Applied Hazard Markings. Where caution, warning, or danger signs or labels are required, the labels must meet the following:
(1) The markings must use words, colors, or symbols that effectively warn personnel (Fig. 2).

Note: ANSI Z535.4, Product Safety Signs and Labels, provides guidelines for the design and durability of signs and labels.
(2) The label can’t be handwritten, and it must be permanently affixed to the equipment (Fig. 3).

Exception to (2): Labels that contain information that’s likely to change can be handwritten, if it’s legible.

Author’s comment: A permanently affixed sign would include a sticker but not a piece of paper taped to the equipment.

(3) The marking must be of sufficient durability to withstand the environment involved.

Analysis: In an effort to standardize the marking requirements found in the NEC, this requirement has been added. Certainly most will agree that a typed sign is better than a handwritten one. There are several changes to the Code that require compliance with this section, but even without those changes this rule still applies. For example, 110.16 includes a statement requiring compliance with this section. If that statement weren’t in 110.16, it wouldn’t make this rule optional. Some of the locations in which this rule applies are 110.16, 110.22, 250.20 (although there’s no reference in that section) 312.8, 404.6, 408.3, and many others.

Change No. 3: Spaces about Electrical Equipment

Changes have been made to the door hardware requirements for large equipment, and outdoor dedicated electrical space provisions have been added to this edition.

Article 110 Requirements for Electrical Installations
Part II. 600V, Nominal, or Less
110.26 Spaces About Electrical Equipment
(C) Entrance to and Egress from Working Space.
(3) Personnel Doors. If equipment with overcurrent or switching devices rated 800A or more is installed, personnel door(s) for entrance to and egress from the working space located less than 25 ft from the nearest edge of the working space must have the door(s) open in the direction of egress and be equipped with listed panic hardware (Fig. 4).

Author’s comments: History has shown that electricians who suffer burns on their hands in electrical arc flash or arc blast events often can’t open doors equipped with knobs that must be turned. Because this requirement is in the NEC, the electrical contractor is responsible for ensuring that panic hardware is installed where required. Some electrical contractors are offended at being held liable for nonelectrical responsibilities, but this rule is designed to save the lives of electricians. For this and other reasons, many construction professionals routinely hold pre-construction or “pre-con” meetings to review potential opportunities for miscommunication — before the work begins.

Analysis: The requirements for door swings and hardware in equipment rooms containing large equipment are among the most important safety provisions for electricians in the entire NEC. When an electrical accident occurs — particularly an arc flash or arc blast — the victim is rendered incapable of opening a door if the door swings in or has hardware that must be operated. Because the victim’s hands and eyes are the two parts of the body most often affected, this issue is even more important. These incidents occur at currents much less than 1,200A. By decreasing the current threshold of this requirement to 800A, more electrical rooms will be equipped with outward swinging doors with panic hardware.

The requirements for the door hardware itself have also changed. Since the inception of this rule in 2002, the hardware requirements for the outward swinging door have allowed for panic hardware or similar devices that open without the victim having to manipulate a knob or similar hardware. In this new edition of the Code, there’s no longer any option. Panic hardware must be used, and it must be listed. Most people will agree that a safety feature like panic hardware probably should be a listed piece of equipment, but now that’s the only option. Hardware like a push/pull plate is no longer allowed, despite the fact that it’s even easier to use than the required panic hardware.

(E) Dedicated Equipment Space. Switchboards, panelboards, and motor control centers must have dedicated equipment space as follows:
(2) Outdoor. Outdoor installations must comply with 110.26(E)(2)(a) and (b).
(a) Installation Requirements. Outdoor electrical equipment must be installed in suitable enclosures and be protected from accidental contact by unauthorized personnel, by vehicular traffic, or by accidental spillage or leakage from piping systems.
(b) Dedicated Electrical Space. The footprint space (width and depth of the equipment) extending from the floor to a height of 6 ft above the equipment must be dedicated for the electrical installation. No piping, ducts, or other equipment foreign to the electrical installation can be installed in this dedicated footprint space (Fig. 5).

Analysis: New to this edition of the NEC is a requirement for outdoor equipment to have dedicated space above and below it, similar to the requirements for indoor installations. The intent of the dedicated space rule for indoor equipment is to allow the installer enough space to install raceways and cables. This is just as important outdoors as it is indoors.

Change No. 4: Multiwire Branch Circuits

The requirement for grouping multiwire branch circuit neutral conductors has been lessened.

Article 210 Branch Circuits
Part I. General Provisions
210.4 Multiwire Branch Circuits.
(D) Grouping. The ungrounded and neutral conductors of a multiwire branch circuit must be grouped together by cable ties or similar means at the point of origination (Fig. 6).

Exception: Grouping isn’t required where the circuit conductors are contained in a single raceway or cable unique to that circuit that makes the grouping obvious, or if the conductors have circuit number tags on them (Fig. 7).

Author’s comments: Grouping all associated conductors of a multiwire branch circuit together by cable ties or other means within the point of origination makes it easier to visually identify the conductors of the multiwire branch circuit. The grouping will assist in making sure that the correct neutral is used at junction points and in connecting multiwire branch circuit conductors to circuit breakers correctly, particularly where twin breakers are used. If proper diligence isn’t exercised when making these connections, two circuit conductors can be accidentally connected to the same phase or line.

Caution: If the ungrounded conductors of a multiwire circuit aren’t terminated to different phases or lines, the currents on the neutral conductor won’t cancel, which can cause an overload on the neutral conductor (Fig. 8).

Analysis: The requirement for grouping the neutral of a multiwire branch circuit was added in the 2008 edition of the NEC. It allows for cable ties or “similar means.” This “similar means” is the subject of debate among many. Is electrical tape “similar means?” Is identifying the neutral conductor with numbered tags, corresponding to the ungrounded conductors, “similar means?” With this change, it becomes clear that placing circuit number tags on the neutral conductors is a suitable method and can be used instead of cable ties.

Change No. 5: GFCI Protection

The requirements for GFCI protection have been greatly expanded.

Article 210 Branch Circuits
Part I. General Provisions
210.8 GFCI Protection. Ground fault circuit interruption for personnel must be provided as required in 210.8(A) through (D). The ground fault circuit-interrupter device must be installed at a readily accessible location.
(A) Dwelling Units. GFCI protection is required for all 15A and 20A, 125V receptacles installed in the following locations:
(7) Sinks. GFCI protection is required for all 15A and 20A, 125V receptacles located within an arc measurement of 6 ft from the outside edge of a sink.

Analysis: Previous editions of the Code required that, in a dwelling unit kitchen, only the receptacles serving the countertop needed to be GFCI protected. For all other sinks, most notably wet bar sinks, the receptacles within 6 ft of the sink needed to be protected. Consider, for example, a disposal receptacle under the sink. If we’re discussing a wet bar (that isn’t in a kitchen), GFCI protection will be required. If, however, the wet bar is in a kitchen, protection isn’t required because it isn’t a countertop receptacle. The safety concern is exactly the same in both installations, so why were the rules different? They’re more similar now with this change, although kitchens will still have more GFCI requirements than wet bars. Kitchens now require GFCI protection for the countertop receptacles [210.8(A)(6)] and for receptacles within 6 ft of the sink.

(9) Bathtubs or Shower Stalls. GFCI protection is required for all 15A and 20A, 125V receptacles located within 6 ft of the outside edge of a bathtub or shower stall.

Analysis: GFCI protection for 125V, 15A and 20A receptacles in bathrooms is required by 210.8(A)(1). Although it’s rare, there are instances where a shower or bathtub is located remote from the other plumbing fixtures and therefore isn’t in the bathroom. In cases such as this, GFCI protection has never been required, despite having (more or less) the same dangers as a bathroom. Obviously, both bathrooms and shower rooms involve water, but they also might involve tile floors, which significantly increase the risk of electrical shock. Although this change doesn’t specify the fact that it applies only to showers and bathtubs that aren’t in a bathroom, 210.8(A)(1) still requires GFCI protection for all 15A and 20A, 125V receptacles in a bathroom — not just those within 6 ft of the shower or bathtub.

(10) Laundry Areas. All 15A and 20A, 125V receptacles installed in laundry areas of a dwelling unit must be GFCI protected (Fig. 9).

Analysis: Nobody can argue the fact that GFCI protection has significantly decreased the number of electrocutions in the United States. Despite having no examples of deaths or injuries to point out in this Code change proposal, the NEC has now incorporated laundry areas into the list of locations requiring GFCI protection. In previous editions of the Code, this was only required if the receptacle was within 6 ft of a sink.

When the laundry sink rule first came into the NEC, many people were nervous, thinking that a washing machine wouldn’t work on a GFCI-protected circuit (which, of course, it will). This new requirement for laundry areas was submitted by one of the leading manufacturers of appliances in the United States, so perhaps that will silence those who think GFCIs and motors don’t play well together.

It’s worth noting that the Code uses the term “laundry area,” not “laundry room.” If the laundry equipment is in a very large laundry room, perhaps the AHJ will only require the GFCI protection within 6 ft of the washing machine. Maybe they’ll require protection for the entire room. As always, when in doubt, check with the inspector if you think this rule is going to create a problem in the field.

(B) Other than Dwelling Units. GFCI protection is required for all 15A and 20A, 125V receptacles installed in the following commercial/industrial locations:
(3) Rooftops. All 15A and 20A, 125V receptacles installed on rooftops must be GFCI protected (Fig. 10).

Author’s comment: A 15A or 20A, 125V receptacle outlet must be installed within 25 ft of heating, air-conditioning, and refrigeration equipment [210.63].

Exception 1 to (3): Receptacles on rooftops aren’t required to be readily accessible other than from the rooftop.

Analysis: Section 210.8 requires that GFCI devices discussed in this section must be readily accessible, which means that they are capable of being reached quickly by those who need to access the equipment. Climbing over or removing obstacles — or having to use a portable ladder — also violates this rule [Art. 100 accessible, readily]. In a commercial building, there doesn’t need to be a permanent ladder to the roof unless the roof is more than 16 ft in height (refer to the International Mechanical Code). A rooftop without a permanent ladder obviously requires a portable ladder, and is therefore not readily accessible, according to Art. 100.

Because GFCI devices must be installed in readily accessible locations [210.8], a rooftop without a permanent ladder can’t have a GFCI device. While that doesn’t sound like a major issue at first, you probably wouldn’t want to be the person on the roof who has to go on an Easter egg hunt through the building trying to find the reset button.

Interestingly, it could be argued that this exception isn’t required at all. A close read of the definition says that those “to whom ready access is requisite” aren’t required to use a portable ladder. If a GFCI is on the roof, then who needs the ready access? It will probably be the person who’s already on the roof to begin with. It will be he or she that might trip the GFCI, and they’re the ones who need access to it. It can even be argued that having a GFCI in the building instead of on the roof violates this rule because you’re now forced to use a portable ladder to get back down and reset it!

(8) Garages. All 15A and 20A, 125V receptacles installed in garages, service bays, and similar areas must be GFCI protected, unless they’re in show rooms or exhibition halls (Fig. 11).

Analysis: The 2011 version of the Code added requirements for commercial garages, but only in the areas where electrical diagnostic equipment, electrical hand tools, or portable lighting were being used. The rules for GFCI protection are typically governed by location, not by specific equipment usage. GFCIs are required in commercial garages because the floor is conductive, not because “electrical diagnostic equipment, electrical hand tools, or portable lighting” are more hazardous than other electrical equipment. By removing this text from the rule, GFCI protection is now required for things like engine block heaters, which are very common in some parts of the country. These block heaters aren’t always listed. Therefore, they may have more leakage current than product standards allow, which greatly increases the possibility of electrocution. Eliminating the list of equipment also helps the enforcement community in a major way, as inspectors are no longer required to try to figure out what each receptacle in the garage is for and enforce GFCI requirements based on what is often complete guesswork.

(D) Dwelling Unit Dishwashers. Outlets supplying dishwashers in a dwelling unit must be GFCI protected (Fig. 12).

Analysis: No one will argue that GFCIs reduce the likelihood of a fatal electrical shock and have greatly reduced the number of electrocutions since their inception. New to this edition, we have a rule for dwelling unit dishwashers. This change was warranted, according to the submitter of the proposal because of the track record of GFCIs and the data showing the decrease in electrocutions in this country. According to some members of the Code-Making Panel, “any electrical appliance combined with the use of water should be GFCI protected.” It may seem odd that this rule isn’t found in 210.8(A) since that’s the location for dwelling units. The reason that it’s found here in (D) is because (A) only applies to receptacle outlets, whereas this new rule requires GFCI protection for dwelling unit dishwashers, even if the equipment is hardwired.

GFCI protection has also been added to several other Articles in the NEC.

A new requirement regarding the accessibility of GFCI devices has been added for appliances.

Article 422 Appliances
Part I. General
422.5 Ground-Fault Circuit-Interrupter (GFCI) Protection. GFCI devices required in this Article must be readily accessible.

Analysis: In order to provide consistency with other Code sections, such as 210.8, this section has been added. By ensuring that the GFCI device is in a readily accessible location, the GFCI can be tested on a monthly basis as indicated in the instructions for the device. I’m not quite sure how the GFCI discussed in 422.51 (which is required to be part of the cord of a vending machine) is going to be readily accessible unless the vending machine is located adjacent to (not in front of) the receptacle serving it. That ought to anger some architects and business owners whose vending machines may now be unplugged by mischievous children or adults!

A new rule requiring GFCI protection for tire inflation and automotive vacuum cleaners has also been added to the Code.

422.23 Tire Inflation and Automotive Vacuum Machines. Tire inflation machines and automotive vacuum machines that are provided for public use must be GFCI protected.

Analysis: Automotive vacuums and tire inflation machines are typically found outdoors, where the person using one will be on concrete, dirt, asphalt, or other surface that can serve as a ground-fault current path. Add in the fact that these surfaces are often covered in rainwater or snow, and you have an electrical accident in the making. This change was added to the Code because of a real-world event involving an electric shock. While a GFCI device may not prevent the shock from happening, it will typically prevent such a shock from being a fatal one.

And last but not least, the requirement for GFCI protection of swimming pool motors has been expanded.

Article 680 Swimming Pools, Fountains, and Similar Installations
Part II. Permanently Installed Pools, Outdoor Spas, and Outdoor Hot Tubs
680.21 Motors.
(C) GFCI Protection. GFCI protection is required for outlets supplying pool pump motors connected to single-phase, 120V through 240V branch circuits, whether by receptacle or by direct connection (Fig. 13).

Analysis: Previous editions of the Code required GFCI protection for swimming pool motors if the circuit supplying the motor was 15A or 20A. It’s pretty tough to argue that the swimming pool should have less protection if it’s fed by a larger circuit (in fact, most will say that it’s impossible to argue that point). Due to this apparent flaw in logic, the NEC has been changed to mandate GFCI protection for the motor regardless of the circuit ampacity, provided that the motor is on a 120V through 240V circuit.

Change No. 6: AFCI Protection

Numerous revisions have been made to the AFCI requirements in the Code.

Article 210 Branch Circuits
Part I. General Provisions
210.12 Arc-Fault Circuit-Interrupter Protection.
Arc-fault circuit-interrupter protection must be provided in accordance with 210.12(A), (B) and (C). AFCI devices must be installed in readily accessible locations.
(A) Where Required. All 15A or 20A, 120V branch circuits in dwelling units supplying outlets or devices in kitchens, family rooms, dining rooms, living rooms, parlors, libraries, dens, bedrooms, sunrooms, recreation rooms, closets, hallways, laundry areas, or similar rooms or areas must be protected by one of the following:
(1) A listed combination type AFCI installed to provide protection of the entire branch circuit.
(2) A listed branch/feeder type AFCI at the origin of the branch circuit, plus a listed outlet branch-circuit AFCI installed at the first outlet box of the branch circuit. The outlet box must be marked to indicate that it’s the first outlet box of the circuit.
(3) A listed supplemental arc protection circuit breaker installed at the origin of the branch circuit, plus a listed outlet branch-circuit type AFCI installed at the first outlet box on the branch circuit. When using this option, the following must be met:
(a) The branch-circuit wiring must be continuous from the branch circuit overcurrent device to the AFCI device.
(b) The maximum length of the branch circuit to the AFCI is 50 ft for 14 AWG conductors or 70 ft for 12 AWG conductors.
(c) The first outlet box in the circuit must be marked.
(4) A regular fuse or circuit breaker plus a listed outlet branch-circuit type AFCI installed at the first outlet of the branch circuit. When using this option, the following must be met:
(a) The branch-circuit wiring must be continuous from the branch-circuit overcurrent device to the AFCI device.
(b) The maximum length of the branch circuit to the AFCI is 50 ft for 14 AWG conductors or 70 ft for 12 AWG conductors.
(c) The first outlet box in the circuit must be marked.
(d) The combination of the branch-circuit overcurrent device and the AFCI must be listed and identified as meeting the requirements for a “System Combination” type AFCI.
(5) A listed outlet branch-circuit type AFCI at the first outlet can be used if the wiring between the overcurrent device and the AFCI contains all metal boxes and is installed using any (or a combination) of the following: RMC, IMC, EMT, Type MC, Type AC cables meeting the requirements of 250.118, metal wireways, or metal auxiliary gutters.
(6) A listed outlet branch-circuit type AFCI at the first outlet of the circuit can be used if the wiring between the overcurrent device and the AFCI is in a raceway with 2 in. of concrete encasement.

Author’s comments: The combination AFCI is a circuit breaker that protects downstream branch-circuit wiring as well as cord sets and power-supply cords; an outlet branch-circuit AFCI (receptacle) is installed as the first outlet in a branch circuit to protect downstream branch-circuit wiring, cord sets, and power-supply cords. The 120V circuit limitation means AFCI protection isn’t required for equipment rated 230V, such as a baseboard heater or room air conditioner.

Exception: AFCI protection can be omitted for an individual branch circuit to a fire alarm system in accordance with 760.41(B) and 760.121(B), if the circuit conductors are installed in metal wireways, metal auxiliary gutters, RMC, IMC, EMT, or steel sheath Type AC or MC cable that qualifies as an equipment grounding conductor in accordance with 250.118, with metal outlet and junction boxes.

Note 3: See 760.41(B) and 760.121(B) for power-supply requirements for fire alarm systems.

Author’s comment: Smoke alarms connected to a 15A or 20A circuit in a dwelling unit must be AFCI protected if the smoke alarm is located in one of the areas specified in 210.12(A). The exemption from AFCI protection for the “fire alarm circuit” contained in 760.41(B) and 760.121(B) doesn’t apply to the single- or multiple-station smoke alarm circuit typically installed in dwelling unit bedroom areas. This is because a smoke alarm circuit isn’t a fire alarm circuit as defined in NFPA 72, National Fire Alarm Code. Unlike single- or multiple-station smoke alarms, fire alarm systems are managed by a fire alarm control panel (Fig. 14).

Analysis: Since the introduction of AFCIs to the NEC in 1999, this section has changed every single Code cycle. Changes have included expanding the locations where they’re required, adding exceptions, addressing existing wiring, as well as extending the wiring, and allowing for devices other than circuit breakers.

The option of using devices other than circuit breakers is no longer written as an exception, but that doesn’t mean that it became any easier to use them. A total of six different options now exist for AFCI protection, including various combinations of circuit breakers and receptacles. Obviously option one is the easiest to follow (and understand), but depending on the prices of the various devices now on the market, it may not be the most economical.

Kitchens and laundry areas have been added to the locations where AFCIs are required. Because a combination AFCI trips on both parallel and series arcs, this seems to make sense. The series arcs that are detected by an AFCI are only those that exceed 5A of current. Because the amount of current in a series arc is limited by the load, few circuits include equipment that will experience this kind of load. Refrigerators and washing machines are some of the first pieces of 120V equipment that come to mind when thinking of energy-consuming appliances, yet they weren’t required to be protected until now.

The NEC now requires AFCI protection for devices as well as outlets. This probably won’t have any real impact, but there have been questions about whether or not a switch that’s located in an area requiring AFCIs needs to have the protection if its load is located in an area that doesn’t require protection (consider a bathroom light whose switch is in the hallway next to the door, which was a common practice in the 1950s). Most people agree that a switch box isn’t an outlet because the switch doesn’t meet the definition of “outlet” as it relates to the “current is taken” portion of the definition. This change removes that issue from the discussion and will force Code geeks to find another point to argue about.

Metal wireways and auxiliary gutters have been added to exceptions one and three, although it’s very doubtful that you’ll ever see a house with an auxiliary gutter.

MC cable is now allowed to be used with concrete encasement as an option for using an AFCI device. It’s probably not often that someone will go to the expense of using a concrete-encased wiring method just to eliminate having to use a breaker, as it hardly seems cost effective; from a safety perspective, there’s no real reason it shouldn’t be allowed.

(B) Branch-Circuit Extensions or Modifications — Dwelling Units. Where branch-circuit wiring is modified, replaced, or extended in any of the areas specified in 210.12(A), the branch circuit must be protected by:
(1) A listed combination AFCI located at the origin of the branch circuit; or
(2) A listed outlet branch circuit AFCI located at the first receptacle outlet of the existing branch circuit.

Exception: AFCI protection isn’t required for extensions less than 6 ft long, as long as there are no outlets or devices added.

Analysis: The requirement for adding AFCI protection to existing branch circuits that are extended received a bit of relief. If outlets or devices are being added as part of the extension of the circuit, AFCI protection is still required. If a person is doing a panel change on an older house, it’s not uncommon for the existing conductors to be a little too short when it’s time to put the new panel in. In these instances, the electrician can add up to 6 ft of conductor length (to each circuit) and not have to burden the homeowner with a few hundred dollar’s worth of AFCI breakers.

(C) Dormitory Units. All 120V, single-phase, 15A and 20A branch circuits supplying outlets installed in dormitory unit bedrooms, living rooms, hallways, closets, and similar rooms must be AFCI protected by one of the methods discussed in 210.12(A)(1) through (6).

Analysis: New to this edition of the Code is subsection (C), which requires AFCI protection for dormitories. Dormitories aren’t defined in the NEC, but if you look up the definition in a dictionary, you will find that is a residence for students. Dormitories have more or less the same hazards as a dwelling unit, so the Code-Making Panel decided to give college kids the same protection that they had growing up at home.

Change No. 7: Conductor Sizing

The rules for sizing branch-circuit, feeder, and service conductors have been simplified.

Article 210 Branch Circuits
Part II. Branch-Circuit Ratings
210.19 Conductor Sizing
(A) Branch Circuits
(1) General. Branch-circuit conductors must have an ampacity of not less than the maximum load to be served. The conductor must be the larger of (a) or (b).
(a) Conductors must be sized no less than 125% of the continuous loads, plus 100% of the noncontinuous loads, based on the terminal temperature rating ampacities as listed in Table 310.15(B)(16), as shown in Fig. 15.

(b) Conductors must be sized to the maximum load to be served after the application of any adjustment or correction factors (Fig. 16).

Exception 1: If the assembly he overcurrent device are both listed for operation at 100% of its rating, the conductors can be sized at 100% of the continuous load.

Author’s comments: Equipment suitable for 100% continuous loading is rarely available in ratings under 400A. See 210.20 for the sizing requirements for the branch-circuit overcurrent device for continuous and noncontinuous loads.

Analysis: The hardest thing to understand in the entire Code book might be how to properly size a conductor, and it seems that every time you get it all figured out you read this section again and realize you’re wrong. This change should go a long way toward clearing up the issue with many end-users.

Article 215 Feeders
215.2 Minimum Rating
(A) Feeder Conductor Size
(1) General. Feeders must be sized to carry the calculated load from Parts III, IV, and V of Article 220. The feeder conductors must be the larger of (a) or (b):
(a) The minimum feeder conductor ampacity must be no less than 125% of the continuous load plus 100% of the noncontinuous load, based on the terminal temperature rating ampacities as listed in Table 310.15(B)(16) [110.14(C)(1)], as shown in Fig. 17.

(b) The feeder ampacity must be not less than the maximum load to be served after the application of any adjustment or correction factors (Fig. 18).

Author’s comment: See 215.3 for the feeder overcurrent device sizing requirements for continuous and noncontinuous loads.

Note 2: To provide reasonable efficiency of operation of electrical equipment, feeder conductors should be sized to prevent a voltage drop not to exceed 3%. In addition, the maximum total voltage drop on both feeders and branch circuits shouldn’t exceed 5%.
Note 3: See 210.19(A), Note 4, for voltage drop for branch circuits.

Exception 1: If the assembly and the overcurrent device are both listed for operation at 100% of their rating, the conductors can be sized at 100% of the continuous load.

Author’s comment: Equipment suitable for 100% continuous loading is rarely available in ratings under 400A.

Exception 2: Neutral conductors can be sized at 100% of the continuous and noncontinuous load.

Analysis: Sizing circuit conductors is, without question, one of the most difficult things to deal with in the NEC. I have personally been involved in e-mail discussions with some of the top Code experts in the country that have been several pages long — with the end result of nobody agreeing on how to size a conductor. This change clarifies the procedure by affirming that the continuous vs. noncontinuous issue (and the terminal ratings) are dealt with in their own calculations, and any ampacity or correction factors (such as the number of current-carrying conductors in a raceway) are dealt with in a separate calculation. The more restrictive (bigger conductor) is then used for the final conductor size.

Additionally, the text regarding conductor sizing for dwellings and mobile homes has been removed from this section, as it’s dealt with adequately in Art. 310.

Article 230 Services
Part IV. Service-Entrance Conductors
230.42 Size and Rating
(A) Load Calculations. Service-entrance conductors must have sufficient ampacity for the loads to be served in accordance with Parts III, IV, or V of Art. 220. The ampacity must not be less than (1), (2), or (3).
(1) 125% of the continuous loads, plus 100% of the noncontinuous loads.

Author’s comment: See 215.3 for the sizing requirements of feeder overcurrent devices for continuous and noncontinuous loads.

(2) The sum of the noncontinuous load plus the continuous load after the application of any adjustment of correction factors.
(3) 100% of the continuous and noncontinuous load if the conductors terminate to an overcurrent device listed for 100% of its rating.

Analysis: This change, like the changes in 210.19 and 215.3, has been made in order to clarify the mathematical procedure used to size conductors. Questions regarding when (in the mathematical process) the adjustment and correction factors are supposed to be factored in have been long unanswered. Hopefully, these changes will fix that glaring problem.

Change No. 8: Lighting Load Calculations

A new exception to the lighting load calculation requirements has been made to more closely reflect reality.

Article 220 Branch-Circuit, Feeder, and Service Calculations
Part II. Branch-Circuit Load Calculations
220.12 Lighting Load for Specified Occupancies. The general lighting load specified in Table 220.12 must be calculated from the outside dimensions of the building or area involved.

Exception: Where the adopted energy code requires a smaller value for lighting loads, the energy code values can be used if the following requirements are met:
(1) A power monitoring system that provides continuous information on the lighting load of the building must be installed.
(2) The power monitoring system must have an alarm that alerts the building owner or manager if the lighting load exceeds the values set by the energy code.
(3) The lighting load demand factors in 220.42 can’t be used.

Analysis: Office buildings require a calculation of 3½VA per square foot, despite the fact that many energy codes don’t allow more than 1½VA to be connected. With the NEC calculation being twice as much as energy codes allow, something has to give eventually. There have been many proposals to the NEC to change this section to reflect energy code requirements, but they’ve all been soundly rejected. This proposal stands out, as it requires a monitoring system to be installed that warns the building owner or manager if the lighting load is exceeding the value that was allowed for with this exception. Members of the Code-Making Panel (CMP 2) had very differing views on this change. Some said this change shouldn’t require any sort of monitoring at all, while others said the change doesn’t take into account task lighting (such as desk lamps) that are often used when energy codes are very restrictive. It should be interesting to see how this evolves over the next few years, as this is the first time an exception like this has been introduced. Perhaps the NEC will include more exceptions like this once there’s a track record of their effectiveness.

Change No. 9: Type of Disconnect

The rule requiring building disconnects to be “suitable for use as service equipment” has been extensively revised.

Article 225 Outside Branch Circuits and Feeders
Part II. Buildings or Other Structures Supplied by a Feeder(s) or Branch Circuit(s)
225.36 Type. The building/structure disconnecting means can be comprised of a circuit breaker, molded case switch, general-use switch, snap switch, or other approved means. If an existing building uses the neutral conductor for the bonding of metal parts [250.32(B) Ex], the disconnect must be listed as suitable for use as service equipment (Fig. 19).

Author’s comment: “Suitable for use as service equipment” means, among other things, that the service disconnecting means is supplied with a main bonding jumper so a neutral-to-case connection can be made, as required in 250.24(C) and 250.142(A).

Analysis: Previous editions of the Code required a switch that was “suitable for use as service equipment” as the disconnect for a building supplied by a feeder or branch circuit. The main difference between a disconnect that’s “suitable for use as service equipment” and one that is not is the ability to connect the neutral to the frame of the disconnect enclosure. This practice is only allowed for buildings supplied by a service [250.24] — not those supplied by a feeder or branch circuit (except for existing buildings discussed in 250.32(B), Exception). Because this Article only deals with branch circuits and feeders, there’s no real need for this requirement. The exception allowing snap switches for residential garages was removed from this rule because now any building supplied by a branch circuit can use a snap switch as a disconnect if it can handle the load.

Change No. 10: Grounding Electrode Conductor Installation

These changes address underground grounding electrode conductors (GECs), as well as buildings supplied by feeder circuits and the requirements for ferrous metal enclosures containing a GEC. In addition, the objects that grounding electrode conductors and bonding jumpers can connect to have also been clarified.

Article 250 Grounding and Bonding
Part III. Grounding Electrode System and Grounding Electrode Conductor
250.64 Grounding Electrode Conductor Installation. Grounding electrode conductors must be installed as specified in (A) through (F).
(B) Conductor Protection. Where installed exposed, grounding electrode conductors must be protected where subject to physical damage and are permitted to be installed on or through framing members. Grounding electrode conductors 6 AWG copper and larger can be installed exposed along the surface of the building if securely fastened and not subject to physical damage.
Grounding electrode conductors sized 8 AWG must be protected by installing them in rigid metal conduit, intermediate metal conduit, PVC conduit, electrical metallic tubing, or reinforced thermosetting resin conduit.

Author’s comment: A ferrous metal raceway containing a grounding electrode conductor must be made electrically continuous by bonding each end of that type of raceway to the grounding electrode conductor [250.64(E)], so it’s best to use PVC conduit or reinforced thermosetting resin conduit.

The cover requirements found in 300.5 don’t apply to grounding electrode conductors or their bonding jumpers.

Analysis: Table 300.5 gives the cover (also known as burial depth) requirements for underground installations. Nothing in that table, or section, indicates whether or not it applies to grounding electrode conductors or bonding jumpers. While it wouldn’t be good to hit one of these conductors with a shovel, you probably wouldn’t get electrocuted.

(D) Grounding Electrode Conductor for Multiple Building or Structure Disconnects. If a service or building/structure disconnect consists of more than a single enclosure, grounding electrode connections must be made in one of the following methods:
(1) Common Grounding Electrode Conductor and Taps. A grounding electrode conductor tap must extend to the inside of each disconnecting means enclosure.
The common grounding electrode conductor must be sized in accordance with 250.66, based on the sum of the circular mil area of the largest ungrounded conductor supplying the equipment (Fig. 20).

A grounding electrode conductor must extend from each disconnecting means, sized not smaller than specified in Table 250.66, based on the area of the largest ungrounded conductor for each disconnecting means.
The grounding electrode tap conductors must be connected to the common grounding electrode conductor, without splicing the common grounding electrode conductor, by one of the following methods:
(1) Exothermic welding.
(2) Connectors listed as grounding and bonding equipment.
(3) Connections to a bus bar of sufficient length and not less than ¼ in. thick × 2 in. wide that’s securely fastened and installed in an accessible location.
(2) Individual Grounding Electrode Conductors. A grounding electrode conductor, sized in accordance with 250.66 based on the ungrounded conductor(s) supplying the individual disconnecting means, must be connected between the grounding electrode system and one or more of the following:
(1) The service neutral conductor
(2) The equipment grounding conductor of the feeder circuit
(3) The supply-side bonding jumper
(3) Common Location. A single grounding electrode conductor is permitted from a common location, sized not smaller than specified in Table 250.66, based on the area of the ungrounded conductor at the location where the connection is made. The grounding electrode conductor must connect the grounding electrode system to one or more of the following:
(1) The service neutral conductor
(2) The equipment grounding conductor of the feeder circuit
(3) The supply-side bonding jumper

Analysis: Previous versions of this Code rule have used the terms “service equipment,” “service disconnect,” “service conductors,” and other service-related terms. By using these terms, the NEC paints itself into a corner by inadvertently excluding buildings or structures that are fed by a feeder circuit. It makes sense that the rules would be the same regardless of the type of circuit feeding the building or structure, as the intent and purpose of the rules are the same in both instances.

The size of the bus bar discussed in this section has also been cleared up. By only providing two of the three dimensions of the bus bar, some Code users were misunderstanding and misusing this section. A typical ground bar that’s found in nearly every panel has two dimensions that are ¼ in. and 2 in. … it just so happens that they’re the wrong dimensions. Considering that this bus bar is connecting multiple grounding electrode conductors and their bonding jumpers, it needs to have considerable mass. This change makes it clear the ground bar in a small panel doesn’t meet the requirements.

(E) Ferrous Metal Enclosures and Raceways Containing Grounding Electrode Conductors.
(1) General. To prevent inductive choking of grounding electrode conductors, ferrous raceways and enclosures containing grounding electrode conductors must have each end of the raceway or enclosure bonded to the grounding electrode or grounding electrode conductor (Fig. 21).

(2) Methods. Bonding must be done by one of the methods discussed in 250.92(B)(2) through (B)(4).
(3) Size. Bonding jumpers must be the same size or larger than the required size of the grounding electrode conductor in the raceway or other enclosure.
(4) Wiring Methods. When a raceway is used for a grounding electrode conductor, it must meet all of the requirements for the raceway, such as securing and
supporting, number of bends, conductor fill, and so forth.

Author’s comment: Nonferrous metal raceways, such as aluminum rigid metal conduit, enclosing the grounding electrode conductor aren’t required to meet the “bonding each end of the raceway to the grounding electrode conductor” provisions of this section.

Caution: The effectiveness of a grounding electrode is significantly reduced if a ferrous metal raceway containing a grounding electrode conductor isn’t bonded to the ferrous metal raceway at both ends. This is because a single conductor carrying high-frequency induced lightning current in a ferrous raceway causes the raceway to act as an inductor, which severely limits (chokes) the current flow through the grounding electrode conductor. ANSI/IEEE 142 — Recommended Practice for Grounding of Industrial and Commercial Power Systems (Green Book) states: “An inductive choke can reduce the current flow by 97%.”

Author’s comment: To save a lot of time and effort, install the grounding electrode conductor exposed if it’s not subject to physical damage [250.64(B)] or enclose it in PVC conduit suitable for the application [352.10(F)].

Analysis: In 2011, 250.64(E) was a remarkably long paragraph. When rules are written in overly long paragraphs, they tend to be misread, misunderstood, or ignored altogether. While there are no real technical changes to the requirements of bonding these raceways, few will argue that the list format the NEC is using more and more often is a better method here.

In addition to the changes noted above, the objects that grounding electrode conductors and bonding jumpers can connect to have also been clarified.

250.68 Termination to the Grounding Electrode.
(C) Grounding Electrode Connections. Grounding electrode conductors and grounding electrode bonding jumpers are permitted to terminate to:
(1) Interior metal water piping located not more than 5 ft from the point of entrance to the building/structure.

Exception: In industrial, institutional, and commercial buildings where conditions of maintenance and supervision ensure only qualified persons service the installation, the entire length of the metal water piping system can be used for grounding purposes, provided the entire length, other than short sections passing through walls, floors, or ceilings, is exposed.
(2) The metal frame of a building/structure.
(3) A concrete-encased electrode can be extended from the concrete to an accessible location above the concrete (Fig. 22).

Analysis: These welcome changes answer many questions Code users have had in the past. It’s important to remember that this section isn’t telling us when structural metal or water pipe is an electrode, as that’s covered in 250.52. This section is simply telling us when we can use these items as conductors to connect other items together. The structural metal of a building may or may not be a grounding electrode, but it’s certainly conductive, so it can be used to connect different electrodes together.

Consider a metal frame of a building that doesn’t meet the definition of a grounding electrode because it doesn’t meet the criteria of 250.52(A)(2). This same building has a water pipe that meets 250.52(A)(1) and therefore must be used as a grounding electrode. The water pipe enters the building 300 ft away from the service disconnect. Is it OK to use the metal of the building as a conductor to connect the water pipe to the service equipment? This will allow me to use a short piece of wire from the service equipment to the structural metal, then walk 300 ft to the water pipe and connect the pipe to the structural metal with another short piece of wire. The answer is now clearly “yes,” as it should be. The structural metal of the building is not only conductive, but it also has low impedance (probably lower impedance than a copper conductor would be, given the size of the metal). Previous editions of the Code only allowed this practice if the structural metal met the criteria of being a grounding electrode.

A new item (3) has also been added to this list, which many people will see as a welcome change. The NEC has been silent on the issue of having a piece of rebar (that meets the criteria of a concrete-encased electrode) exit the concrete and enter the building. That piece of steel couldn’t be called a grounding electrode because that portion of the steel wasn’t in the concrete and therefore didn’t meet the requirements of 250.52(A)(3). This shouldn’t matter, and most installers and inspectors have installed and passed it for years — but now the argument can end.

Change No. 11: Conductor Ampacity

Several changes to 310.15 include removing the dwelling unit table, dealing with the rooftop issue, and clarifying the bundling rules.

Article 310 Conductors and General Wiring
310.15 Conductor Ampacity
(B) Ampacity Table. The allowable conductor ampacities listed in Table 310.15(B)(16) are based on conditions where the ambient temperature isn’t greater than 86°F, and no more than three current-carrying conductors are bundled together.
(3) Conductor Ampacity Adjustment.

(a) Four or More Current-Carrying Conductors. Where four or more current-carrying power conductors are in a raceway longer than 24 in. [310.15(B)(3)(a)(3)], or where cables are bundled for a length longer than 24 in., the ampacity of each conductor must be reduced in accordance with Table 310.15(B)(3)(a) (Table 1).

Author’s comment: Conductor ampacity reduction is required when four or more current-carrying conductors are bundled because heat generated by current flow isn’t able to dissipate as quickly as when there are three or fewer current-carrying conductors (Fig. 23).

(1) Conductor ampacity adjustment of Table 310.15(B)(3)(a) doesn’t apply to conductors installed in cable trays, 392.80 applies.
(2) Conductor ampacity adjustment of Table 310.15(B)(3)
(a) doesn’t apply to conductors in raceways having a length not exceeding 24 in.
(4) Conductor ampacity adjustment of Table 310.15(B)
(3)(a) doesn’t apply to conductors within Type AC or Type MC cable under the following conditions:
(a) The cables don’t have an outer jacket,
(b) Each cable has no more than three current-carrying conductors,
(c) The conductors are 12 AWG copper, and
(d) No more than 20 current-carrying conductors (10 2-wire cables or six 3-wire cables) are installed without maintaining spacing for a continuous length longer than 24 in.
(5) Ampacity adjustment of 60% applies to conductors within Type AC or Type MC cable without an overall outer jacket under the following conditions:
(b) The number of current-carrying conductors exceeds 20.
(c) The cables are stacked or bundled longer than 24 in. without spacing being maintained.

Analysis: The 2011 edition of the NEC made an attempt to clarify the rules for ampacity adjustment as they pertain to spare conductors and other conductors that weren’t clearly addressed in the rule. The attempt didn’t quite work because the title of the table, the table note, and the Code text seemed to argue with each other. By adding the note at the bottom of Table 310.15(B)(3)(a), it is clear that spare conductors are counted for ampacity adjustment, due to the fact that they’ll probably be used eventually, and it would be a difficult sell to a customer to convince them that simply terminating these conductors to equipment suddenly makes every wire in the raceway undersized. In order to avoid that argument, the conductor is counted immediately upon installation.

Conductors that can’t be energized simultaneously are also now discussed in the table. Consider a set of 3-way switches. One switch is fed by the supply conductor while the other switch feeds the luminaire. Between these switches are two conductors, often referred to as “travelers.” Regardless of the position of either switch, one of the travelers will be energized and the other will not. If you change the position (up or down) of either switch, you’ll change which of the travelers is energized, but in no case can you possibly energize both. Due to this, there’s no reason to count both conductors in the ampacity adjustment, as only one of them will be adding heat.

(c) Raceways and Cables Exposed to Sunlight on Rooftops. When applying ampacity adjustment correction factors, the ambient temperature adjustment contained in Table 310.15(B)(3)(c) is added to the outdoor ambient temperature for conductors installed in raceways or cables exposed to direct sunlight on or above rooftops to determine the applicable ambient temperature for ampacity correction factors in Table 310.15(B)(2)(a) or Table 310.15(B)(2)(b), as shown in Fig. 24.

Exception: The ampacity adjustment isn’t required for conductors that are type XHHW-2.

Note 1: See the ASHRAE Handbook—Fundamentals (www.ashrae.org) as a source for the ambient temperatures in various locations.
Note 2: The temperature adders in Table 310.15(B)(3)(c) (Table 2) are based on the measured temperature rise above local climatic ambient temperatures due to sunlight heating.

Author’s comment: This rule requires the ambient temperature used for ampacity correction to be adjusted where conductors or cables are installed in a raceway or cable on or above a rooftop and the raceway is exposed to direct sunlight. The reasoning is that the air inside raceways and cables that are in direct sunlight is significantly hotter than the surrounding air, and appropriate ampacity corrections must be made in order to comply with 310.10.

Analysis: Of all the NEC rules that have been added over the last three Code change cycles, this is one of the most controversial. It started as an Informational Note (actually a Fine Print Note back then), and then grew into a requirement that receives dozens of proposals every three years. New to this edition of the NEC, the conductors in any raceway, not just circular raceways, are required to comply with this requirement. Conductors inside of cable assemblies are also required to comply.

A new Exception for type XHHW-2 has been added as well. Testing showed that this conductor insulation type fared much better than other 90°C conductors as it relates to the impact of high temperatures. Interestingly, this data has not only resulted in the Exception being added, but has also generated a lot of discussion about whether or not the whole “raceways and cables on rooftops” is an issue worth having in the Code at all. There still haven’t been any documented failures to warrant the requirement, and it appears that the Code-Making Panel is finally willing to discuss that fact.

(7) 120/240V, Single-Phase Dwelling Services and Feeders. For one-family dwellings and individual dwelling units of two-family and multifamily dwellings, service and feeder conductors supplied by a single phase, 120/240V system can be sized using 310.15(B)(7)(1) through (4), as shown in Fig. 25.

(1) Service conductors supplying the entire load of a one-family dwelling or an individual dwelling unit in a two-family or multifamily dwelling can have an ampacity of 83% of the service rating.

Author’s comment: 310.15(B)(7) can’t be used for service conductors for two-family or multifamily dwelling buildings.

(2) For a feeder rated 100A through 400A, feeder conductors supplying a one-family dwelling, or an individual dwelling unit in a two-family or multifamily dwelling, can have an ampacity of 83% of the feeder rating, but only if the feeder supplies the entire load of the dwelling.

Author’s comment: 310.15B(7)(2) can’t be used to size feeder conductors where a feeder doesn’t carry the entire load of the dwelling unit.

Warning: 310.15(B)(7) doesn’t apply to 3-wire service or feeder conductors connected to a three-phase, 120/208V system, because the neutral conductor in these systems always carries neutral current, even when the load on the phases is balanced [310.15(B)(5)(b)]. For more information on this topic, see 220.61(C)(1) (Fig. 26).

(3) Feeders for an individual dwelling unit are never required to be larger than the conductors in 310.15(B)(7)(1) or (2).
(4) Neutral conductors are sized using 220.61 and 230.42 for services and 220.61 and 215.2 for feeders.

Caution: Because the service neutral conductor is required to serve as the effective ground-fault current path, it must be sized so it can safely carry the maximum fault current likely to be imposed on it [110.10 and 250.4(A)(5)]. This is accomplished by sizing the neutral conductor in accordance with Table 250.102(C), based on the area of the largest ungrounded service conductor [250.24(C)(1)].

Analysis: There weren’t many NEC rules that received more proposals and public comments than this one. Up until 2008, this requirement didn’t receive much thought during the Code change process, but then it received dozens. Some proposals wanted the allowance deleted altogether while others wanted the paragraph to be written better. The proposals that passed, however, change the allowance so that it only applies when the conductors serve the entire load of the dwelling. This idea generated dozens of proposals and comments in the 2011 edition, due to the fact that it didn’t really make a lot of sense (and it still doesn’t). At the end of the 2011 process, it was decided that there was certainly work that needed to be done, but no one quite had the answer to fix all of the problems. Now, in 2014, the table is gone, and a new 83% reduction has been introduced. Unfortunately, instead of looking at a table to determine ampacity, we now get to break out the calculator as well as the Code book. We’ll have to wait and see what happens in 2017, but it will almost certainly be changing again.

Change No. 12: Securing and Supporting

The allowance for unsupported raceways has been revised, finally, to recognize a commonly accepted practice.

Article 348 Flexible Metal Conduit
Part II. Installation
348.30 Securing and Supporting.
(A) Securely Fastened. Flexible metal conduit must be securely fastened by a means approved by the authority having jurisdiction within 1 ft of termination, and it must be secured and supported at intervals not exceeding 4½ ft.

Exception 1: Flexible metal conduit isn’t required to be securely fastened or supported where fished between access points through concealed spaces and supporting is impracticable.

Exception 2: If flexibility is necessary after installation, unsecured lengths from the last point the raceway is securely fastened must not exceed:
(1) 3 ft for trade sizes ½ through 1¼
(2) 4 ft for trade sizes 1½ through 2
(3) 5 ft for trade sizes 2½ and larger

Exception 4: FMC to a luminaire or electrical equipment within an accessible ceiling is permitted to be unsupported for not more than 6 ft from the last point where the raceway is securely fastened. For the purposes of this exception listed fittings are considered support (Fig. 27).

Analysis: Over the last three Code cycles, several proposals have been made to try to make this allowance make sense. As previously written, a 6 ft length of FMC to a luminaire in a suspended ceiling did, in fact, require support. This is due to the fact that the fitting wasn’t addressed (compare this to the language in 320.30 and 330.30 for AC and MC cables, respectively). Quite often, luminaires come with a 6 ft length of 3⁄8-in. FMC attached to them, commonly known as a fixture whip. These whips, technically, required a support, although few Code users realized it (or cared about it). Most inspectors also either weren’t aware of the rule or didn’t care about it, and have allowed the FMC to go unsupported. This change makes it obvious that a 6 ft length of FMC doesn’t need support when it’s used in a suspended ceiling.

This same change was also made for liquidtight flexible metal conduit (Sec. 350.30) and liquidtight flexible nonmetallic conduit (Sec. 356.30).

Change No. 13: Ampacity of Conductors in Metal Wireways

The requirements for ampacity adjustment in metal wireways were revised in order to create a practical rule.

Article 376 Metal Wireways
Part II. Installation
376.22 Number of Conductors and Ampacity.
(B) Conductor Ampacity Adjustment Factors. When more than 30 current-carrying conductors are installed in any cross-sectional area of the wireway, the conductor ampacity, as listed in Table 310.15(B)(16), must be adjusted in accordance with Table 310.15(B)(3)(a). Signaling and motor-control conductors between a motor and its starter used only for starting duty aren’t considered current carrying for conductor ampacity adjustment.

Analysis: The issue of limiting the number (ampacities) of conductors in a wireway certainly makes sense. The more current-carrying conductors there are in proximity with each other, the more heat there will be. The key word here, however, is “proximity.” A wireway that’s 100 ft long, for example, might contain 31 current-carrying conductors. If these conductors are 10 ft away from any other conductors in the wireway, there certainly won’t be any additional heating involved. It’s where these conductors are in the same cross section that heat becomes a real problem. This change clarifies and solidifies this concept.

Change No. 14: Switch Connections

The requirement for having a neutral at switch locations was revised.

Article 404 Switches
I. Installation
404.2 Switch Connections
(C) Switches Controlling Lighting Loads. Switches controlling line-to-neutral lighting loads must have a neutral provided at the switch location other than in the following locations:
(1) Where the conductors for switches enter the device box through a raceway that has sufficient cross-sectional area to accommodate a neutral conductor.
(2) Where the switch box can be accessed to add or replace a cable containing a neutral without damaging the finish of the building (Fig. 28).

(3) Snap switches with integral enclosures [300.15(E)].
(4) Nonhabitable rooms and bathrooms.
(5) For 3-way and 4-way switches, but only if the entire floor area is visible from the switches’ location(s), as shown in Fig. 29.

(6) Where the lighting is controlled automatically.
(7) Switches controlling receptacles.

Note: The purpose of the neutral conductor is to complete a circuit path for electronic lighting control devices.

Analysis: Once again, when a new, significant rule comes into the Code, it takes at least three more years for the NEC to make it logical. This isn’t the fault of the people writing the Code, it’s just that it’s very difficult to anticipate every type of installation that a new rule will affect. The areas where a neutral isn’t required have been greatly expanded in this edition of the NEC, and the existing exceptions were cleaned up dramatically as well.

The text regarding framing cavities that are “open at the top or bottom” has been removed, as even Code experts disagreed about what that exception meant. It’s been replaced with item two, which is easier to understand and makes more sense as well.

Additional new allowances include 3-way and 4-way switches where the switches can “see” the entire room. In these instances, an occupancy sensor would only be installed at one or the other switch, so there’s no reason to require a neutral at both.

Change No. 15: Grounding and Bonding Transformer Enclosures

Transformers are now required to have a place to terminate conductors.

Article 450 Transformers and Transformer Vaults
Part I. General Provisions
450.10 Grounding
(A) Dry-Type Transformer Enclosures. Where separate equipment grounding conductors and supply-side bonding jumpers are installed, a terminal bar for these conductors must be installed inside the enclosure. The terminal bar must not cover any ventilation openings (Fig. 30).

Exception: Where a dry-type transformer is equipped with wire-type connections (leads), the terminal bar isn’t required.

Analysis: When you consider the number of green and bare wires that are typically found in a transformer, it’s really a wonder that this rule hasn’t been in the Code for 100 years. The primary conductors supplying the transformer will usually contain an equipment grounding conductor. The secondary conductors will include a supply-side bonding jumper. There will (typically) be a system bonding jumper inside the transformer. There will also be (typically) a grounding electrode conductor in the transformer. Wouldn’t it be nice if there was a place to terminate all these wires?

Change No. 16: Receptacles and Attachment Plugs in Class I and II Locations

These changes make it clear that receptacles installed in Class I and Class II locations must be part of the building wiring.

Article 501 Class I Locations
Part III. Equipment
501.145 Receptacles and Attachment Plugs, Class I, Divisions 1 and 2
(A) Receptacles. Receptacles must be part of the premises wiring, unless they’re part of a temporary portable assembly [501.140(A)(5)], as shown in Fig. 31.

(B) Attachment Plugs. Attachment plugs must provide for the connection of an equipment grounding conductor and be identified for Class I locations.

Author’s comments: See the definitions of “Attachment Plug” and “Receptacle” in Art. 100. Receptacles listed for Class I locations can be any of the following types:

– Interlocked switch receptacle. This receptacle contains a built-in rotary switch interlocked with the attachment plug. The switch must be off before the attachment plug can be inserted or removed.

– Manual interlocked receptacle. The attachment plug is inserted into the receptacle, and then it’s rotated to operate the receptacle switching contacts.

– Delayed action receptacle. This receptacle requires an attachment plug and receptacle constructed so that an electrical arc will be confined within the explosionproof chamber of the receptacle.

Article 502 Class II Locations
Part III. Equipment
502.145 Receptacles and Attachment Plugs. Receptacles and attachment plugs must be identified for the location in which they’re being used.
(A) Class II, Division 1.

(1) Receptacles. In Class II, Division 1 locations, receptacles must be part of the premises wiring.

(2) Attachment Plugs. Attachment plugs must provide for connection of the circuit equipment grounding conductor of the flexible cord.
(B) Class II, Division 2.
(1) Receptacles. In Class II, Division 2 locations, receptacles must be part of the premises wiring.
(2) Attachment Plugs. Attachment plugs must provide for connection of the circuit equipment grounding conductor (Fig. 32).

Analysis: The use of makeshift assemblies of cords, plugs, and receptacles is never a great idea, but when used in a hazardous location it creates a deadly situation. Hazardous locations are classified as such because, as the name implies, they’re more hazardous than a typical location. The introduction of a spark in one of these locations can have catastrophic results, including fire and explosion. By ensuring that receptacles used in a hazardous location are part of the premises wiring, the likelihood of these accidents occurring is greatly reduced.

Change No. 17: Wiring Methods in Special Occupancies

Copper equipment grounding conductors are no longer required for agricultural buildings, marinas and boatyards.

Article 547 Agricultural Buildings
547.5 Wiring Methods
(F) Separate Equipment Grounding Conductor. An equipment grounding conductor for circuits in agricultural buildings must be insulated or covered when installed underground (Fig. 33).

Analysis: After studying the corrosive effects of the various items found in agricultural buildings (methane gas, ammonia, carbon dioxide, hydrogen sulfide, etc.), it was determined that aluminum conductors perform as well as copper conductors. With this new information, the Code now allows aluminum equipment grounding conductors to be used.

Article 555 Marinas and Boatyards
555.15 Grounding
(B) Type of Equipment Grounding Conductor. An insulated equipment grounding conductor is required for circuits.
(C) Size of Equipment Grounding Conductor. The insulated equipment grounding conductor must be sized in accordance with 250.122, but not smaller than 12 AWG.

Analysis: It seems that aluminum still has the negative stigma attached to it that developed in the 1960s and 70s. Not only have installation practices improved since then, but the conductors themselves have as well. New and better alloys are used in the manufacture of aluminum conductors, and they have an excellent track record. Furthermore, the effects of sea water and freshwater on aluminum conductors are minor. With this change, aluminum equipment grounding conductors are now allowed.

Change No. 18: Disconnecting Means for Electric Signs and Outline Lighting

The location of the sign disconnect is now addressed.

Article 600 Electric Signs and Outline Lighting
Part I. General
600.6 Disconnects. Each circuit for a sign, outline lighting system, or skeleton tubing must be controlled by an externally-operable switch or circuit breaker that will open all ungrounded conductors. Where the circuit is a multiwire branch circuit, the switch or circuit breaker must open all ungrounded conductors of the circuit simultaneously.
(A) Location.
(1) At Point of Entry to a Sign Enclosure. The disconnect must be located at the point where the conductors for a sign or outline lighting system enter a sign enclosure or pole, and it must disconnect all conductors of the circuit (Fig. 34).

Exception: The disconnect isn’t required when the conductors are in a raceway and pass through the sign.

(2) Within Sight of Sign. The disconnecting means must be within sight of the sign or outline lighting system. If the disconnecting means is out of the line of sight from any section of the sign or outline lighting able to be energized, the disconnecting means must be lockable, as described in 110.25 (Fig. 35).

Author’s comment: According to Art. 100, “within sight” means that it’s visible and not more than 50 ft from one to the other.

(3) Within Sight of the Controller. Signs or outline lighting systems operated by electronic or electromechanical controllers located external to the sign or outline lighting system must have the disconnecting means installed in accordance with (1) through (3):
(1) Located within sight of or in the same enclosure with the controller.
(2) Be capable of disconnecting the sign or outline lighting and the controller from all ungrounded supply conductors.
(3) Be lockable, as described in 110.25.

Analysis: Although signs have had disconnect requirements for quite some time, its physical location has been a gray area of the Code. By ensuring that the disconnect is installed at the point that the conductors enter the sign, installers can rest assured that all of the conductors inside the enclosure are de-energized. According to the submitter of this proposal, who represents the International Sign Association, sign disconnects are often installed in odd locations, which results in some of the conductors in the enclosure remaining energized with the switch in the OFF position. This change should enhance worker safety.

Change No. 19: Outdoor Spa and Hot Tub Installations

The bonding requirements for outdoor spas and hot tubs has been revised, once again, and wiring method requirements have been expanded.

Article 680 Swimming Pools, Fountains, and Similar Installations
Part IV. Spas and Hot Tubs
680.42 Outdoor Installations. Electrical installations for outdoor spas or hot tubs must comply with Parts I and II of this Article, except as permitted in (A) or (B).
(B) Bonding. Bonding is permitted by mounting equipment to a metal frame or base. Metal bands that secure wooden staves aren’t required to be bonded. Bonding of perimeter surfaces [680.26(B)(2)] isn’t required for spas and hot tubs if the spa or hot tub meets all of the following:
(1) Listed as a self-contained spa for aboveground use.
(2) Not identified as suitable only for indoor use.
(3) Installed in accordance with the manufacturer’s instructions and located on or above grade.
(4) The top rim must be at least 28 in. above all perimeter surfaces that are within 30 in. measured horizontally from the spa or hot tub. Nonconductive external steps for entry to the spa can’t be used to reduce or increase the rim height measurement.

Analysis: The equipotential bonding requirements for an outdoor spa or hot tub have been puzzling, at best, for a long time. Changes to 680.42(B) were made in an effort to clarify what the rules are for these pieces of equipment. The bonding requirements for spas and hot tubs are the same as the requirements for swimming pools, other than the perimeter surface allowances that were added by this change. The common installation of an aboveground spa or hot tub doesn’t result in any real danger from nearby conductive surfaces, unless the top of the spa or hot tub is less than
28 in. from the surface. When it’s that low, a person can have one foot in the water and another on a concrete (or similar) floor and be in a dangerous predicament should the water be (or become) energized.

(C) Interior Wiring for Outdoor Spas or Hot Tubs. Any Chapter 3 wiring method containing a copper equipment grounding conductor insulated or enclosed within the outer sheath of the wiring method and not smaller than 12 AWG is permitted in the interior of a dwelling unit for the connection to motor, heating, and control loads that are part of a self-contained spa or hot tub, or a packaged spa or hot tub equipment assembly (Fig. 36). Wiring to an underwater light must comply with 680.23 or 680.33.

Analysis: Subsection (C) was modified to allow two-family dwelling units to take advantage of the allowances of this subsection. If the spa or hot tub is installed in a location other than a dwelling unit, the requirements of 680.21(A) requiring RMC, IMC, PVC, EMT, or Type MC cable listed for the location apply.

Change No. 20: Emergency Illumination

A new requirement for emergency illumination of some electrical equipment has been added.

Article 700 Emergency Systems
Part IV. Circuits for Lighting and Power
700.16 Emergency Illumination. Emergency lighting systems must be designed and installed so that the failure of any individual lighting element, such as the burning out of a lamp, won’t leave in total darkness any space that requires emergency illumination.

Author’s comment: This means that a single remote head is never sufficient for an area. A minimum of two lighting heads is always required. This is why individual unit equipment (sometimes called “lunch boxes” in the field) always has two lighting heads.

When an emergency system is installed, emergency illumination must be provided to illuminate the service or building disconnecting means, if the disconnect is located indoors (Fig. 37).

Analysis: This new requirement was put into the Code so that an injured electrician can be located by emergency response personnel. Many of us in the electrical industry have seen pictures or videos of an unfortunate electrician who was the victim of an arc flash or arc blast at the service or building disconnect. When these incidents occur, the main disconnect is probably going to trip, resulting in a severely injured electrician laying in complete darkness waiting for emergency response. With this change, the first responders will not only have the lighting required to find the victim, but will also have lighting to perform whatever type of medical aid the incident warrants.