A new NEC is on tap, with some of the most far-reaching changes ever made as we begin the next 100 years Code. Part 2 of a 4-part series.

This analysis of the changes represented in the 1996 edition of the National Electrical Code (NEC) covers only the most significant of the many changes. Some of those changes were indeed major, including seven new articles, (with another article deleted), and others were for very minor editorial reasons. Major or minor, if you are involved with the particular subject matter, the most minor change can assume major proportions in unexpected ways. Every proposal was made for a reason, and this magazine article is no substitute for careful study of the Code itself. This month's issue covers from Art. 364 through the end of Chapter 4.

Sec. 364-8. The rules for making branches from busways have been overhauled into two subsections, (a) on permitted Chapter 3 wiring methods, and (b) on flexible cord or cable exits. This latter subsection, in the basic rules, tracks the former requirements in this section, as well as that in Sec. 400-8 and Ex. However, the new subsection includes a new exception, which is a dramatic departure in allowable wiring practices.

EC&M tip: Bus drop cable and flexible cord are completely different entities for enforcement purposes. Flexible cord is covered in Art. 400 and is finely stranded to facilitate frequent movement; bus drop cable is in Art. 364 and is more coarsely stranded because it is only intended to connect to stationary equipment.

What happened: Subsection (a) on Chapter 3 extensions from busways only consists of editorial changes (for example, instead of naming the wiring methods, the article titles are used instead) from the previous NEC. The action is in subsection (b).

The basic rules in (b) apply to both cords approved for hard usage and listed bus drop cables used to make flexible connections to portable or stationary equipment. The horizontal travel of the cord or cable cannot exceed 6 ft, and strain relief must be provided at the busway plug-in device and at the equipment termination. The cord or cable must hang vertically from the horizontal support, which must be approved.

The new exception applies against the 6-ft restriction, but only in industrial occupancies with qualified supervision. In this case the "cord or cable" can extend any distance from the plug-in device on the busway. The cord or cable must be supported at least every 8 ft. The other rules still apply. The only change is in the allowable distance, which could be literally anywhere on the plant floor. This is a direct conflict with Sec. 400-8 Ex., which establishes a 6-ft maximum horizontal travel requirement.

The only way out of the conflict is to limit the application of the new exception to bus drop cable, which is not within the scope of Art. 400. Although this may not have been intended, if you apply the Code in its entirety, you may wire under the exception at any distance from the busway, but only if you are using listed bus drop cable to connect to stationary equipment. Check with the AHJ if this will pose a problem.

Background: There are large industrial production areas where this type of flexibility could be very advantageous due to frequent repositioning of large equipment; indeed, the proposal came from General Motors. It does, however, raise serious questions about this type of wiring becoming a de facto substitute for permanent wiring in the building. As such, it will be controversial in many areas.

Sec. 370-16. This section has been completely rewritten to simplify the box sizing rule. Simply compare the size of the box [from Sec. 370-16(a)] with the volume allowances of what you need to put in the box [Sec 370-16(b)], and judge the box accordingly.

What happened: The two drawings itemize the essential aspects of box fill calculations. In the process of reorganizing this section, some substantive changes were made, as follows:

* In Sec. 370-16(a)(2) the commas previously omitted from the former Section 370-16(b) around the words "and nonmetallic boxes" have been restored, so as to clarify that the marking rule applies to nonmetallic boxes.

* Sec. 370-16(b)(2) spells out that the clamp allowance for factory supplied integral clamps (see drawing) still applies, just the same as it does for removable clamps in a metal box, even if the volume of the box were measured with the clamps. The rule is for the factory to measure the box, all boxes, without any clamps, and then in the field you must always provide a clamp allowance for one or more clamps if they will be used.

* In Sec. 370-16(b) the controversy over counting small fittings as a full box fitting allowance has been put to rest; locknuts and bushings don't count.

* In Sec. 370-16(b)(5) each set of equipment grounding conductors (normal and isolated) are now treated alike (a single volume allowance for each based on the largest equipment grounding conductor in the group).

* Sec. 370-16(b)(1) Ex. for fixture wires in a canopy has been changed in three ways. First, the former second sentence of the exception, which only covered full-sized pigtails extending from the box to a canopy, has been deleted because it is unnecessary. Second, a canopy must now be domed; a flat canopy (cover?), even if provided with the fixture and thinner than the cover requirement in Sec. 370-41, no longer qualifies under the exception. Third, similar canopies can make use of the exception. Although definitely intended to be included, a ceiling paddle fan canopy is not literally a "fixture" canopy.

* Finally, an asterisk footnote to the phrase "Maximum Number of Conductors" in Table 370-16(a) was added as follows: "*Where no volume allowances are required by Section 370-16(b)(2) through 370-16(b)(5)." This correlates with the reorganization of the section.

Background: This section originated in the distant past when the majority of wiring was done in metal boxes with metal cables and raceways. At that time it made sense to organize the section around assumed conductor capacities, with occasional deductions for unusual events (other than the routine device deduction). No one does this today. You add up everything that will count, and then compare it with the size of the box.

Note that the new wording will increase the volume ratings in some nonmetallic boxes. Previously test labs have been measuring boxes with integral clamps exactly as supplied, with the clamps still in place. Then, they have said that a further volume allowance in the field didn't need to be taken for such clamps. This was a misinterpretation of the rule and needlessly penalizes the volume rating of the box. Now all boxes, metal and nonmetallic, will be eventually measured the same way, without any clamps. Integral clamps will be milled out to measure volume. Then in the field, whenever these boxes have been (or are) used with internal clamps, a clamp allowance must be applied.

Always remember that figuring volume allowances in a box is far less precise than it appears. Just because box volume can be measured to a hundredth of a cubic inch, that doesn't mean that anything else is so precise. The volume taken up by clamps changes according to how thick the cable(s) are within them, and how many clamps are used. Yet the rule is only a single allowance for all clamps collectively. The volume occupied by a conductor depends on its length, and yet no rule (other than the 6-in. minimum in Sec. 300-14) dictates how long each wire is left. Devices notoriously differ in total size, and yet the allowance is the same for all.

The change in the canopy exception is another good example, where now such canopies need to be "domed." That term is dimensionless because any appreciable volume at all will reasonably, if not exactly, accommodate fixture wire volume. For example, a 4-inch canopy with a 1/4-inch dome has a volume over 3 [in.sup.3]. The rule is a reasonable compromise based on extensive field experience.

Sec. 370-23(e). A new exception (No. 2) has been added that provides comprehensive requirements for cantilevered fixtures.

EC&M tip: Always keep in mind that the basic objective of these requirements is to ensure that excessive force will not be applied to the conduit threads at the supply end of the conduit.

What happened: The new exception includes many specific requirements that ensure that the worst-case mechanical advantage against the conduit threads on the supply end is 3:1, and that 60 lbs is the heaviest static load, as follows:

* The conduit arm (rigid or intermediate metal conduit only) must be unbroken.

* The conduit must be securely fastened with the maximum extension not exceeding 3 ft.

* The minimum length on the supply side of the support must be 1 ft.

* The conduit arm and fixture must not be where passers-by might easily hit or hang on them. Unless inaccessible to unqualified persons, the minimum height is 8 ft, and the minimum horizontal reach is 3 ft from balconies, etc. Some billboards qualify for bottom-mounted fixtures by having fences that restrict unqualified access.

* No fixture supported by a single conduit can exceed 12 in. in any direction from the point of conduit entry. This reduces wind loading and minimizes possible torque on the conduit threads at the fixture end.

* No single conduit can carry more than 20 lbs; with the 3:1 worst case leverage, the highest static load on the supply end comes out 60 lbs.

* The conduit must be threaded wrenchtight into the box (or conduit hub); a connector, such as a set-screw connector cannot be used.

Background: The basic rule requires that boxes that carry fixtures and use the entering raceways for support must be supported by two or more conduits threaded in wrenchtight and secured within 18 in. of the box. A cantilevered box obviously doesn't come close to meeting several of the essential elements of the rule, and yet such installations are common at storefronts and billboards. This change puts clear limits on where these arrangements are acceptable.

Note that the requirements stop at the outlet. If a listed fixture incorporates a conduit stem, then that stem will have been evaluated by the test lab. Also note that Sec. 300-15(d) recognizes wiring compartments in fixtures as the equivalent of boxes. If such a fixture enclosure has (or is equipped with) a threaded conduit hub, then it falls within these requirements. Finally, if two (or more) conduits support the fixture, the size (and weight) can go up, such as long fluorescent fixtures, for example.

Sec. 370-23(g). New rules cover pendant conduit stems that support fixtures. ([ILLUSTRATION OMITTED] on page 46.)

EC&M tip: This rule is a companion to the cantilevered box rule, and is very similar. There are important differences, however, because as long as there is a flexible connection, which is feasible for a pendant, you aren't likely to break the conduit threads.

What happened: Sec. 370-23(g) now has two halves, the first for pendant boxes on cords, for which the rules didn't change, and the second for conduit stems. The rules for conduit pendants are as follows:

* The pendant must be rigid or intermediate steel conduit, threaded wrenchtight into the box or fixture wiring compartment, or into conduit hubs identified for the purpose.

* If the stem is over 18 in. long, the connection at the supply end of the conduit must be flexible.

* If there is only a single conduit involved, the conduit connection at the box must not be easily or inadvertently loosened. This can be done by adding set-screws to the conduit joint, or by installing the fixture so its lowest point is at least 8 ft above the floor or standing platform, and so it is at least 3 ft horizontally from a reaching exposure.

* The fixture must not exceed 12 in. in any horizontal dimension, if there is only a single support. Note that there is no weight limit in this case, beyond the 50-lb general limit for all fixtures supported from boxes in Sec. 410-16(a).

Background: As noted, this is a companion change to the cantilevered box rule. Note that this rule also applies to exit signs, which are listed as exit fixtures by testing labs. The allowance for set screws to prevent loosening of the conduit joints was taken from Sec. 501-9(a)(3), which covers pendant fixtures in Class I Div. 1 locations.

Sec. 374-9(e). This subsection, as in other places in Art. 374, has been rewritten to cover indoor and outdoor uses of both metallic and nonmetallic auxiliary gutters.

EC&M tip: Be sure to track the requirements for multiple conductor derating penalties when using nonmetallic auxiliary gutters and wireways. The absence of the 30 conductor waiver from Note 8(a) can make a big difference.

What happened: Sec. 374-9(e)(1) now covers sheet metal auxiliary gutters and simply requires that in wet locations they be "suitable for such locations." Sec. 373-9(e)(2) covers nonmetallic auxiliary gutters, and is further divided into two requirement groupings, the first for outdoor use and the second for indoors.

All nonmetallic auxiliary gutters must be listed for the maximum ambient temperature of the installation, and also marked for the installed conductor insulation temperature rating. In addition, they must have expansion fittings installed if the expected movement from thermal expansion and contraction will exceed 1/4 in.

If the installation is outdoors, a nonmetallic auxiliary gutter must comply with those requirements, and in addition, must be listed and marked as suitable for wet locations and also as suitable for exposure to sunlight.

Background: Auxiliary gutters are constructed in the same way as wireways, but they differ in their use. They are not raceways. The drawing shows the auxiliary gutter clearly supplementing the wiring spaces in the panelboard end gutters, and not functioning as an ongoing raceway. Normally they do carry a joint listing as wireways, however. After the nonmetallic wireway provisions of Art. 362 (Part B) came into the Code, this was a natural follow-on inclusion.

Sec. 384-4. This section has been completely rewritten. All four of the previous FPNs have been deleted, along with two of the four previous exceptions. The result is much easier to understand.

EC&M tip: Don't confuse the functions of the two zones. The first is for conduit access to the equipment, whether needed or not. The second is workspace and must remain clear of all equipment, even electrical equipment.

What happened: This complete rewrite makes few substantive changes, but is much more clear to the user.

* The various zones are in separate paragraphs so to clarify that there is no need to dedicate the space above working clearance zone. This confusion generated the former FPN No. 4, so that could be dropped.

* The term "rooms" was dropped. As a practical matter, as long as the Code allows "rooms or spaces" (as it did), what we always end up with are, in fact, spaces. This rewrite put an end to the sophistry involved in relabeling electrical rooms as utility rooms so some other utility can share the room. There really is nothing wrong with this as long as there is no safety problem with activities adjacent to the electrical equipment.

* The former exception for ventilating, heating, and cooling equipment (Ex. 2) has been dropped. This was in the Code in order to allow an electrical room to be conditioned. It was not intended to allow this equipment to infringe on the specifically dedicated areas immediately adjacent to the electrical installation. Since this section now covers requirements for the specifically dedicated spaces and not entire rooms, there is no longer any need for the exception.

* The industrial exception (former Ex. 3) has been retained, as well as former Ex. 1. They have been located to follow the rules under exception; old Ex. 1 applies generally and follows the general rule, and the industrial exception that allows relief on equipment over the switchboard or panelboard with appropriate protection from leakage, etc. follows the dedication rule on the width and depth of the equipment. All other fine print notes and exceptions have been folded into the text.

* One final note concerns the new phrase "and motor control centers" in the opening wording. This directly correlates with Section 430-1, exception No. 1.

Background: According to NFPA staff, the electrical department routinely got more questions on this section of the Code than practically any other. The rewrite is much clearer and should help put an end to those questions.

Sec. 410-4(d). This has been expanded to include track lighting and ceiling fans as being excluded from the zone around a tub.

What happened: The section has been renamed, from "Pendants" to "Above Bathtubs." The exclusion zone remains at 3 ft measured horizontally around the tub, and up to the 8-ft level measured above the tub rim, not the floor. In addition to pendants, hanging fixtures, and other cord-connected fixtures, lighting track is now excluded, even if no fixtures are installed within the specified zone. In addition, all ceiling fans are excluded! Although the only intent was to exclude paddle fans, the actual wording excludes all fans in the ceiling.

Background: The idea on the fan was to put a paddle fan without a light on the same footing as the same fan with a light kit installed. On the prior Code, the second fan could be excluded, but the first one had to be allowed. The change corrects that inconsistency. Unfortunately the actual wording reached a large group of fans that were never intended to be included, for obvious reasons. Also note that this rule only applies around standing water; there are no such restrictions in a shower stall.

EC&M tip: Many, if not the majority of ceiling exhaust fans are located with a 3-ft radius of the tub, since the shower is the principal source of moist air. Make sure you reach an understanding with the AHJ as to where these "ceiling fans" will be allowed locally in view of the new wording.

410-56(c). The color orange, by itself, is no longer the essential identifying feature of isolated-ground (IG) receptacles. In addition, it is now a direct Code violation to ground an isolated-ground receptacle directly to the outlet box, effectively canceling its isolated-grounding features. Also, IG receptacles in nonmetallic boxes must now use nonmetallic faceplates.

EC&M tip: Although you cannot cancel the IG aspects of an IG receptacle, the Code does not say how far upstream the isolation needs to be maintained. That is still a design question, and depends on how the ground references for the area served have been engineered. In some cases isolation as far as the local branch-circuit panelboard is sufficient because the sensitive equipment is all related to one room; in other cases the isolation needs to be maintained all the way to the building disconnect.

What happened: There were three changes to this subsection:

* Orange, by itself, no longer is an IG receptacle identifier. The appropriate identifier is an orange triangle, which can still (as in the past) be used on an orange receptacle face.

* An IG receptacle must be installed as such; it must be arranged to maintain isolation between the receptacle equipment grounding terminal and the wiring system equipment grounding conductor. The distance upstream isn't specified as long as the initial grounding is only to an isolated grounding conductor in accordance with Sec. 250-74 Ex.4.

* If you install an IG receptacle in a nonmetallic box, now you must use a nonmetallic faceplate.

Background: The orange triangle. Changing the identification on IG receptacles means that now they can be made in any color, as long as there is an orange triangle visible on the face. This also frees up the color orange (with no triangle) to be used as yet another color of a receptacle for whatever design reasons are thought necessary, perhaps to indicate a connection to a standby system. This does not mean that orange receptacles will no longer be produced as IG receptacles. If the manufacturer so chooses, he may continue to do so. However, (per Code) what actually distinguishes the IG feature is the triangle with the orange color within it, even if it isn't in a contrasting color.

IG receptacles to be so installed. IG receptacles are supposed to be for that purpose, and not just to remind passers-by of pumpkins. No one does this just to spend a lot of extra money on orange devices. From time to time, however, an installer will think he has an IG system based on a totally erroneous choice of wiring methods. When this happens, the practice now prohibited by this change in the Code may show up extensively.

For example, one entire new building, specified as having IG receptacles throughout, was wired with interlocking-armor Type MC cable with only one grounding conductor. Remember, the armor of interlocking-armor Type MC cable isn't recognized as an equipment grounding return path. Wiring that grounding conductor exclusively to the receptacle terminal meant that the entire wiring system was disconnected from a meaningful equipment grounding reference.

When the inspector rejected the job on the finish inspection, the contractor bonded the metal boxes to the IG grounding conductors. This created, in effect, plain orange receptacles out of the premium quality IG receptacles that were specified. Although not exactly unsafe, this practice is at best unethical and would now end.

Nonmetallic faceplates. Because IG receptacles have no connection between their grounding terminals and yokes, they must be mounted in a grounded metal box in order for their yokes to be grounded. The grounded yoke is often the only practical way to ground a metal faceplate, which is a requirement of Sec. 410-56(d). The result is that normally you can't put a metal faceplate on an IG receptacle if it is in a nonmetallic box.

All of this has been true for many years. Apparently there were numerous violations, however, because now Sec. 410-56(c) has a new sentence expressly forbidding the use of a metallic faceplate if an IG receptacle is in a nonmetallic box, even if, as in the case of multigang boxes, there is provision for grounding a metal faceplate.

There is another problem, one this rule shares with Sec. 380-12 (requiring metal faceplates for snap switches to be grounded), namely, the absence of practical enforcement after the inspector leaves. Normally untrained persons freely exchange nonmetallic for metallic faceplates without any permitting process. This is particularly true in residential occupancies most commonly wired with nonmetallic boxes.

Sec. 410-57. Receptacles must no longer be installed in a bathtub or shower space.

This puts receptacles on the same footing as switches, that were excluded from these areas in Sec. 380-4 in the 1993 NEC. A receptacle in such a location is asking for trouble, no matter how well grounded and GFCI protected. This is a new subsection (c), with the other subsections relettered accordingly. Note that this wording, excluding receptacles from the tub or shower space, is somewhat broader in coverage (depending on interpretation) than the Sec. 380-4 exclusion, which is only from wet locations in these areas.

Sec. 410-102. The required track load has been decreased, and now is supposed to apply only to feeder calculations.

What happened: This section on track loading now covers "load calculations" instead of "branch-circuit calculations." In addition, the required allowance has been decreased from 180VA per 2 ft to 150VA per 2 ft. Finally a new FPN states that the VA rating "is for load calculations only and does not limit the length of track that can be run or the number of fixtures allowed."

Background: The intent is to allow greatly increased flexibility in track layouts, particularly where very long runs use only a few fixtures at relatively low loads, as in museums. There are problems, however. The only change in the mandatory rule was to convert "branch-circuit" to "load." In addition, the second sentence of the same rule clearly applies to branch circuits. However, branch circuits are equally subject to load calculations, as covered in Part A of Art. 220. Sec. 220-3(c)(5) refers back to this section for these calculations, and furthermore, no correlating change was made in Art. 220 Part B on feeders.

EC&M tip: The FPN is very clear, but, as a note, it cannot be literally enforced. Check with the AHJ to see whether the new wording will be applied as intended.

Art. 411. This is a new article on low-voltage lighting systems, "and their associated components." These are, in effect, defined as follows: "A lighting system consisting of an isolating power supply operating at 30 volts (42.4 Vpk) or less under any load conditions with one or more secondary circuits, each limited to 25A maximum, supplying lighting fixtures or associated equipment identified for the use." These lighting systems must be listed for the purpose.

These systems have been in use for many years in landscape applications, trade shows, and other places of assembly. These rules set the basic parameters of these systems, as follows:

* Not over 30V (or not over the peak voltage equivalent of 30V RMS ac, 42.4V);

* Isolating power supply required;

* 25A maximum size for all circuits on the secondary side;

* The systems, as a whole, must be listed;

* They must not penetrate ("concealed or extended through a building wall") unless using Chapter 3 wiring methods;

* The systems must not run within 10 ft of pools, spas, and fountains, etc., "except as permitted by Art. 680." Although Sec. 680-6(b)(1) generally allows fixtures to be as near as 5 ft to a pool, Sec. 680-6(b)(2) requires GFCI protection for fixtures between a 5-ft to 10-ft radius unless the fixtures are rigidly attached to a structure at a point at least 5 ft above the maximum water level. Since the secondary circuits need to be ungrounded and isolated, it is unclear whether the GFCI rules properly apply at all. The safest course is to stay 10 ft away, or 5 to 10 ft away but above the 5-ft level and rigidly attached to a structure.

* The secondary circuits must be ungrounded and insulated from the primary supply by an isolating transformer. Exposed bare conductors are permitted under certain conditions. Bare conductors are permitted at least 7 ft above the finished floor unless specifically listed for a lower height.

* The systems must be supplied from a 20A (max.) branch circuit. Art. 411 does not give a maximum voltage, however, Sec. 250-5(a)(1) sets the maximum voltage to ground on the primary where this class of voltage (30V ungrounded) is supplied at 150V. Therefore, as a practical matter, these systems will be supplied from 120V circuits.

Sec. 422-8(d)(1 and 2). These rules on kitchen waste disposers, dishwashers, and compactors have been expanded to allow cord connections in nondwelling occupancies.

The words "intended for dwelling use" have been removed in both places. This will add important design flexibility in commercial occupancies.

Sec. 426-28. Resistance electric deicing and snow-melting equipment alone must now have protection against low-level arcing ground faults through ground-fault protection of equipment (GFPE) on their branch circuits. The former requirement in Sec. 426-53, which applied throughout the entire article, has been relocated to the end of Part C of the article as new Sec. 426-28.

In the former location, the rule applied to such items as skin effect heating in Part E of the article, which was completely impractical. Typical trip values that won't nuisance trip on resistance circuits (including the newer self-regulating cables) run in the 30mA range, although for short runs a conventional GFCI (4-6mA) may hold. Capacitive leakage current in long runs may nuisance trip these devices. Note that the literal text refers to protecting the branch circuit and not the outlet, and therefore a GFPE circuit breaker with the appropriate trip setting would be preferred. Note also that this is branch-circuit level protection; GFPE on a 2000A service disconnect somewhere ahead of this installation is not the required protection.

Sec. 427-22. GFPE is now required on all pipe tracing applications, not just those using cable without a metallic covering. A new exception, conditioned on ground-fault sensing and alarms, provides relief in qualified industrial occupancies where continuity of heat is essential for safe operation.

EC&M tip: This GFPE is neither the very large equipment for main feeders and services 1000A and up, nor is it the 4-6mA GFCI devices that protect against electric shock. Most circuit breaker manufacturers now make small, branch-circuit style breakers in the 15-30A sizes with 30mA GFPE trip settings. To clarify this point, the requirement is now for GFPE to "be provided for each branch circuit supplying..."

What happened: This section has required GFPE on heat tracing circuits for many years, but only if there wasn't a metal jacket. This change removed the words "Not having a metal covering," which then meant that all of these installations required GFPE.

An industrial occupancy with qualified maintenance and supervision and equipment or processes for which continued circuit operation is necessary, may forego the automatic trip of a GFPE circuit breaker. However, the facility must provide for "alarm indication of ground fault" if this option is chosen.

Background: The expansion of GFPE coverage resulted in part from the development and widespread availability of 30mA trip devices in conjunction with very favorable industrial experience. 1000 Ft of Type MI cable will run about 8mA of capacitive leakage, far short of what would nuisance trip one of these devices. For those with critical processes who aren't convinced, the exception may be useful, but if that alarm goes off, qualified personnel must respond immediately. Low level arcing faults on these heating cables almost never trip a conventional overcurrent device, and many fires have resulted.

Sec. 427-23. Effective July 1, 1996, all resistance electric heating equipment within the scope of Art. 427 must have a grounded metal covering, bath on heating wire and cables [new subsection (a)], and also on heating panels [new subsection (b)]. In either case, the metal covering must provide an effective ground return path.

This is a major change. Even though Sec. 427-22 is changing to require GFPE on all these circuits, they must still be produced with a grounded shield around the cable assemblies, on the side away from the heated vessel in the case of heating panels.

The Code is now assuming that even a metal pipeline may be useless at functioning as a ground return path. The grounded covering should immediately return enough current to make the GFPE protection open. In addition, the shield will help shunt currents to ground caused by personnel inadvertently cutting into the cable. This can happen where the cable is under a heavily insulated pipe. The delayed effective date should allow manufacturers to deplete old stock.

Sec. 430-7(a)(8). Either the locked-rotor amperes or the code letter is now permitted on motor nameplates.

This is part of an ongoing process to convert from the code-letter basis for sizing components to the NEMA design letter.

Sec. 430-7(a). The NEMA design letter far Design B, C, D, or E motors must now be marked on the motor. A new FPN notes the related standards, such as NEMA MG1, that define the design designations.

The new required marking appears as a new item (9) on the list, and the remaining items (9) through (13) become items (10) through (14).

Sec. 430-7(b). The code letter has been replaced as the basis far short-circuit and ground-fault protective device ratios in Table 430-152.

EC&M tip: In general, the new Design E motors are the higher efficiency motors. The process of shifting over to design letters instead of code letters has produced a significant simplification of the rules in this article.

What happened: The mandatory language in this section that required code letters to be used in that table (second paragraph, former second sentence), has been deleted. This is a companion action to the other revisions in this section.

Background: Code letters for motors first appeared in the 1940 NEC, so they have been with us a very long time. However, with the new motor designs, they no longer really worked. One result is actually a significant simplification of Table 430-152.

Sec. 430-22(a). The conductors between a wye-start/delta-run controller and the motor must be based on 58% of the motor full-load current.

What happened: A new third paragraph to Sec. 430-22(a) clarifies the requirements for selecting conductors for these motors, each of which carry a portion of the full-load current (FLA / 1.73 = 0.58 x FLA). Note that the phrasing and sentence structure on the line and load side of the controller are very similar.

* Line side: "The selection "shall be based on the motor full-load current."

* Load side: "The selection" shall be based on 58 percent of the motor full-load current."

EC&M tip: Don't confuse the expression "based on" with the final result. In the case of the line side of the controller, the conductors will usually end up at 125% of the full-load current, however, if duty cycle service is involved that percentage could vary greatly. Similarly, in the case of the conductors going to the motor, the usual result will be 73% of the full-load current (58% x 1.25 = 72.5%). But again, in the case of duty cycle service, that result will vary.

Background: Wye-delta starting cuts the locked-rotor starting current to one third the usual value, and is very useful on loads with long acceleration times or for loads with comparatively low starting torque requirements such as fans. It has been an important industrial mainstay for reduced-voltage starting because no special equipment is required to vary system voltages or frequencies; all that is required is a motor with all twelve (or six for a single voltage) leads brought out.

There are no changes to the required ground-fault and short-circuit protection required for these motors; that is still based on the full-load current. There is precedent elsewhere in Art. 430 for this type of consideration. These conductors could be considered a form of tap. Sec. 430-53(d) on single-circuit taps in a multimotor installation allows such taps (at least one-third of the branch circuit ampacity, and not over 25 ft long). In this case the "tap" conductors will be 58% of the parent conductors. There will be complete overload protection for each, but there is no maximum length specified. On balance this has been viewed as an acceptable trade-off over the many years these systems have been used.

Sec. 430-22(a). A new third exception covers conductor sizing to adjustable speed drive equipment. These conductors must now be sized at 125% of the rated input.

Prior to this change, the only code language on this topic was the general requirement in Sec. 430-2 that the "incoming branch circuit...shall be based on the rated input..." The usual inference was to size the conductors at 100% of the rated input. That has now changed with the definitive new language here. Once again, the Code is making a distinction between "based on" and the final result, similar to wye-delta motor conductors.

Sec. 430-52. The section has been completely reorganized with all paragraphs numbered. The short-circuit and ground-fault protective device for any motor must be capable of carrying the starting current of the motor. With the exception of torque motors [requirements unchanged, except now in subsection (d)], all motor fault-protective devices must comply with the applicable numbered paragraphs under Sec. 430-52(c).

In addition to the section reorganization, there are numerous changes in this section, some critical, and all as detailed in the following material. The citations are from the 1996 NEC, with 1993 equivalents in brackets.

Sec. 430-52(c)(1) Ex. 1[-(a) (first paragraph) Ex. 1]. The first numbered paragraph [(c)(1)] requires that a protective device must be selected with a rating or setting that doesn't exceed the value calculated in accordance with the provisions of Table 430-152. Ex. 1 covers the procedure to follow when that calculation results in a size or rating that doesn't correspond to standard size of overcurrent protective device. You are now permitted to round upward to the next higher standard sized device by right, without showing that the next lower size would be unable to carry the load.

What happened: If the Table 430-152 calculation doesn't come out on a standard size, you are now allowed to round upward to the next higher size. The process is similar to Sec. 240-3(b), although there is no 800A upper limit on this exception as there is in Art. 240.

Background: It is difficult to understate the practical significance of this particular code rule to both professional designers as well as those working in the trade. "How do you size an overcurrent device for a motor circuit?" appears, in some form, on every test for electrical licensure and inspector certification administered in this and other countries. The question always was, do you round down, or do you round up?

For code cycle after cycle prior to the 1993 NEC, the code making panel responsible for this material did two things consistently. They absolutely insisted that the exception was a "round-down" exception in panel documents, and that the next higher size was only available in unusual circumstances. They also refused to write a rule that unambiguously stated that objective, and they even rejected proposals that squarely stated what they insisted the rule intended.

The result was increasing chaos, as various jurisdictions began to believe what the panel wrote in their documentation about the intent of the rule and rounded down. Since the exception was ambiguous, either interpretation could be supported. Meanwhile, the overwhelming majority of the country left the panel in their fools' paradise, happily rounding up because the exception had every appearance of being the same as the one in Sec. 240-3.

In 1993, this all changed, as the panel finally rewrote the exception in a form that could not be misinterpreted. Calculations were to be rounded down unless the resulting size wouldn't carry the load. This generated a counter-reaction that resulted in the current text, which is equally unambiguous in the direction of allowing rounding up. This may just settle the question once and for all.

As a result of the change, small motors whose required protection calculated below 15A can round up to 15A. It isn't necessary to consider restricting the overcurrent protective devices to small fuses in this case, which was an inadvertent result of the 1993 NEC. Be aware, however, of product listing restrictions, as covered in Sec. 430-52(c)(2).

Sec. 430-52(c)(3) Ex. 1 [-(a) (second paragraph) Ex.]. For Design E motors, the upper limit on an instantaneous trip circuit breaker can now be as high as 1700% of rated full-load motor current.

What happened: Design E motors now have a special allowance to account for their higher starting inrush currents of 1700%, up from 1300% for all other motors. The wording of the exception also incorporates the revisions in the Table 430-152 standard calculations for motors generally. Other motors now can use 800% instead of the former 700% by right, and Design E motors can now use 1100% by right. Therefore, threshold values in the exception have risen accordingly.

Background: High-efficiency motors with high inrush currents are a fact of life due to market forces responding to energy costs, and also as a result of government regulations. There has been a great deal of attention paid to how much current these motors use, particularly during locked rotor conditions when the current is largely controlled by the resistance of the windings. This resistance has been decreased with the new designs to result in less [I.sup.2]R energy waste, with the side problem of high inrush.

Great care needs to be taken when short-circuit and ground-fault protective devices are set as high as now allowed. A 30 hp 460V motor (Table FLA = 40A) could now be wired on a circuit with protection as high as 680A. A 200 hp 600V motor could be protected as high as 3200A. Although this will probably allow the motors to start, (and indeed, as long as these currents are confined to the motor windings there isn't any problem) this setting may exceed the current flow in an arcing ground fault.

If that happens at 277V (or 347V or 600V) to ground, the fault will only be cleared by the running overload protection. Arcing faults at this voltage tend to restrike and continue, and over the course of the time delay built into running overload protection, the result will be an almost certain fire. There is no substitute for a very high standard of workmanship and careful attention to Sec. 250-51. You must do everything possible to ensure the lowest possible impedance in any ground return path, so any fault returns enough current to exceed the trip setting of the circuit breaker.

Sec. 430-83(a). A new exception (listed as No. 1, forcing the renumbering of all existing exceptions accordingly) has been added covering ratings for Design E motor controllers. These controllers can be either marked for this use, or be derated versions of conventional controllers.

What happened: The new mandatory exception affects Design E motor controllers for motors over 2 hp. When you select a controller for one of these motors, you have two options. The first is to find a controller that has been marked for service with a Design E motor and with a horsepower rating not less than that of the motor in question. The second option is to use a conventionally rated controller with a horsepower rating that exceeds that of the motor in question by a factor of 1.4 from 3 hp to 100 hp, and by a factor of 1.3 for larger motors.

Background: Design E motors have higher locked rotor currents, running about eight times full-load current instead of the conventional six times. The rules for equipment that must be capable of interrupting stalled rotor currents (controllers and disconnects) reflect this by shifting the requirements for using conventional equipment on Design E motors by approximately the ratio of 8:6, or 1.3+. You can also use equipment that has been tested for the higher Design E currents and so marked.

Sec. 430-109. A new exception (listed as No. 1, forcing the renumbering of all existing exceptions accordingly) has been added covering ratings for Design E motor disconnects. These disconnects can be either marked for this use, or be derated versions of conventional disconnects.

The new mandatory exception affects Design E motor disconnects for motors over 2 hp. When you select a disconnect for one of these motors, you have two options. The first is to find a disconnect that has been marked for service with a Design E motor and with a horsepower rating not less than that of the motor in question. The second option is to use a conventionally rated disconnect with a horsepower rating that exceeds that of the motor in question by a factor of 1.4 from 3 hp to 100 hp, and by a factor of 1.3 for larger motors. Please refer to the discussion on Design E controllers at Sec. 430-83(a) for more details.

Table 430-152. This table has been extremely simplified, as all references to code letters have been deleted.

What happened: All polyphase motors are now treated alike, except for synchronous or wound rotor motors. Design E motors now have special percentage for instantaneous trip circuit breakers [please refer to the discussion under Sec. 430-52(c)(3)]. In addition, all ac motors have an increase in the instantaneous trip column from 700% to 800%, with Design E singled out for 1100%. "High reactance squirrel-cage" motors have been deleted because NEMA reported that motor manufacturers didn't know what they were. DC motors no longer distinguish between over 50 hp and 50 hp and below, with the old values for the 50 hp and smaller sizes now applying to all of them. Class CC fuses go into the table as a footnote indicating they should be treated as nontime-delay fuses [please refer to the discussion under Sec. 430-52(c)(1) Ex. 2a.]

[TABULAR DATA FOR TABLE 430-152 (REVISED) OMITTED]

Background: This table now has approximately the same degree of complexity as it did in the 1937 NEC, just before the advent of Code Letters. Now, with the increasing use of different motor design classes, the old code letter concept has lost much of its relevance. Note, however that Table 430-7(b), setting the parameters for locked-rotor currents, has been retained for use with older motors.

Sec. 440-14. In qualified industrial occupancies where hermetic compressors are used for other than air-conditioning, the local disconnect at the compressor may now be omitted. The controller disconnect must be able to be locked in the open position.

What happened: A new exception (inserted as No. 1; existing exception becomes No. 2) has been added that allows the normally required disconnect at refrigeration equipment to be omitted in industrial occupancies with qualified maintenance and supervision. The controller disconnect must be capable of being locked open, and the compressor must not be used for air-conditioning.

Background: The effect of this new exception is to put these compressors back on the same footing as motors generally. The local, in-sight disconnect can be omitted if the controller disconnect can be locked open. Traditionally, this section has imposed a more severe requirement on this equipment because those servicing the equipment often don't have access to an upstream disconnecting means. Under the conditions described in the exception, this isn't likely to be a problem.

There are industrial occupancies that use very large refrigerant compressors for process cooling. Occasionally, this equipment is in hazardous (classified) locations, and it commonly operates at medium voltages. Requiring a local disconnect under these conditions is often impractical and could introduce additional hazards.

Sec. 450-23. On outdoor installations, less-flammable liquid-insulated (LFLI) transformers must now comply with Sec. 450-27 unless they are installed near Type I or Type II buildings and all conditions of the liquid listing are satisfied.

What happened: This section has been completely reorganized into separate subsections, (a) for indoor installations and (b) for outdoor installations. There are no substantive changes in (a), but (b) has been changed to reference Sec. 450-27 in cases where the listing conditions of the LFLI transformer fluid cannot be met or if the transformers are on, adjacent to, or on the roof of other than Type I or II construction. If there are high-hazard areas in the transformer areas [ILLUSTRATION OMITTED], a new FPN directs the reader to Sec. 450-27 for additional safeguards, but the status as a FPN makes enforceability questionable.

Background: The concept behind the change, besides readability, is to coordinate the requirements for LFLI and mineral oil transformers so the transformers with the comparatively flammable fluid (mineral oil) don't end up with the most relaxed requirements. For the traditional concrete noncombustible construction (Type I or Type II), adherence to the liquid listing is sufficient. If this cannot be done, then Sec. 450-27 rules apply.

The FPN indicates, however, that there may be higher hazards in areas such as adjacent combustible material in storage, or an area near a critical fire escape. In such a case, additional safeguards such as those cited in Sec. 450-27 may be necessary, even where the main rule would normally allow the transformers in accordance with the liquid listing. What is not clear is how they would be enforced due to the unenforceable status of FPNs.

EC&M tip: The relative degree of hazard is necessarily subjective to a considerable degree. Be sure to discuss the issues with the AHJ well ahead of time.