Feeders to hot tubs

Q:Our county inspectors now require us to run the feeders to an outdoor hot tub subpanel in a listed raceway from the main service equipment at the meter; unless we run the feeders to the load center supplying the house's branch circuits in conduit. In other words, if we ran SER cable from the main disconnect to the load center, the feed to the hot tub would not be allowed to come from this load center. It would have to have a disconnect tapped at the main service equipment and then be run in conduit with the proper ground wire. Does this sound right? As far as I'm concerned, if your main feeders are run to your load center and if you come from this panel with the approved raceway and proper ground wire, it should be OK. Your advice would be greatly appreciated, as we work on many large custom homes are running circuits several hundred feet from the service equipment.

A: Your inspector is correct in requiring the feeder from the service to be in one of the raceways allowed by Section 680-25(d). If you have used Type SER cable to feed the house panel, the circuit for the hot tub can't originate in the remote panelboard. Of course, if this is an existing home with a hot tub added later, Section 680-25(d)(1) allows the panelboard to supply the hot tub — provided the equipment grounding conductor in the cable is insulated.

If the panel for the hot tub is located in a separate building, you could apply Section 680-25(d)(2). In this case you must also comply with Section 250-32 and Part B of Article 225.

It is too early to make positive statements about the 2002 revisions to the NEC. However, it does look as if we will see an extensive revision of Article 680. When the 2002 edition is published you may see some relaxed rules covering an installation of this kind.
Dann Strube

Neutral conductors

Q: We are doing a service change-out at a local manufacturing plant. The service is two 4,000A 480/277V three-phase, four-wire mains with a tie-breaker. Two new 2,500kVA pad-mounted transformers feed the mains. The existing service is the same (very old equipment). Almost all the load is three-phase, three-wire. There is some 277V lighting in the plant. We measured the current neutral amperage at 100A. We are planning to use 11 500 MCM THHN for phase-feeder conductors. We want to use 11 350MCM THHN for the neutral conductor. I believe that this is allowed by the 1999 NEC in Section 250-24(b)(1) and (2). Our local inspector wants us to make the neutral conductor the same size as the phase conductors. He said the Code requires the neutral to be able to handle the short-circuit current.

A: Section 250-24(b) applies to the service grounded conductor, so I assume you are talking about the service conductors and not a feeder. A feeder would have an equipment grounding conductor to handle ground fault currents, and that conductor would be sized from Table 250-122 based on the rating of the overcurrent device.

The grounded service conductor, which is a neutral in your case, must be sized for both normal neutral currents and ground fault currents. The sizing for normal neutral off-balance currents is covered by Section 220-22. This must include not only the measured current you describe, but the maximum off-balance current. This would be based on the 277V loads only since 480V three-wire loads will not contribute to neutral current.

The sizing for ground-fault currents (one of the types of fault currents, but not exactly the same as short-circuit currents) is covered by Section 250-24(b)(1) and (2). These rules require the grounded conductor to be sized according to Table 250-66, but because the total area of your conductors is 11 × 500 = 5,500 kcmil, the minimum size for a single conductor would be 5,500 by 12.5% = 687.5 kcmil. However, if you are running the 11 sets of 500 kcmil in parallel raceways, 250-24(b)(2) requires that the grounded conductor in each raceway be based on Table 250-66, and not smaller than No. 1/0. For a set of 500 kcmil conductors, Table 250-66 requires a 1/0 conductor. The ampacity of 11 sets of 1/0 (assuming normal conditions of use) would be 11 × 150 = 1,650A, and the total circular mil area of these 11 conductors would be 11 × 105.6 = 1,162 kcmil. Your 350 kcmil conductors would be more than adequate for both the neutral loads and the fault currents.
Noel Williams

Surge Protection

Q: I am working on a project installing metering equipment at a remote natural gas pipeline station. My problem is the telephone company has installed their customer-interface junction box inside of a Class 1, Div. 2 classified area.

Inside the box is their lightning surge protector. I first determined that this was acceptable per NEC 501-14(b)(3), which applies to communication circuits and doesn't even require that the surge protector be a solid-state non-arcing type.

Next, though, NEC 501-17(b) requires that the surge protectors be either non-arcing or sealed. I assume that the prior code applies since it is specifically for communication circuits. To top it off, NEC 800-30(c) states that a primary protector must not be located in a classified location. This seems contradictory unless the protector at the station entrance is not considered a primary protector.

Also, the NEC seems to use the word, protector,; at times to mean a surge protector, and other times to mean an overcurrent protector or fuse.

A: Although it appears that there are conflicts between the sections you referenced, these sections are actually coordinated with each other. As you concluded, Section 501-14(b)(3) is the primary rule in your situation. This rule is specific to communications protectors in Class I, Div. 2 areas and permits ordinary protectors in ordinary enclosures in such areas. Although Section 800-30(c) prohibits protectors in classified areas in the main rule, the exception to this rule recognizes Section 501-14. Section 501-17 refers to surge arrestors, which are the types of equipment covered by Article 280. Surge arrestors, primary protectors and transient voltage surge suppressors (TVSS) are different products.

Surge arrestors are covered by ANSI/IEEE C62.1, C62.2, C62.11 and C62.22. Primary protectors are covered by ANSI/UL 497 “Standard for Protectors for Paired Conductor Communications Circuits.” TVSS devices are covered by UL 1449 and NEMA LS 1. TVSS devices are the subject of a new Article 285 in the 2002 NEC.

Although all of these are sometimes called protectors; they are separate products and are referred to by different names in the NEC. In fact, there are a number of other types of protectors; for specific uses, such as protectors for fire alarm circuits, but these are also covered by separate product standards.

I find that protectors; can be clearly distinguished (at least in the NEC) by their full names: primary protectors; secondary protectors; thermal protectors; short circuit protectors; and overcurrent protective devices
Noel Williams

Minimum gauges for branch circuits

Q I could not find in the Code anything requiring #12 AWG wire as the minimum size used for branch circuits in commercial work.

I also believed there was an NEC rule that said #14 AWG is permitted for control functions in a commercial application. My inspector said there is nothing in the codes about the gauge minimum.

A: The 1999 NEC specifies the minimum size of branch circuit conductors. Section 210 19(d) gives the minimum size as #14 AWG. The same requirement is found in Section 310-5. Section 310-5 also contains nine exceptions and covers voltages up to 35KV.

For remote control and signaling work the requirements are in Sections 725-27(a) and 725-51 Exception. In these sections the smallest conductor allowed is #18 AWG.

You may find local codes that require a minimum size of #12 AWG. However, the NEC sizes are as given above.
Dann Strube

Shore-tie receptacles

Q: We are doing a dock project in which we have 400A shore tie receptacles for ships. We have four 400A circuits.

Each 400A circuit has two 400A receptacles, a 480V three-wire receptacle and one 480V four-wire receptacle. This seems to conflict with NEC 555-6. Do you agree?

A: I see no direct conflict with Section 555-6, which is concerned with load calculations for feeders and services that supply shore power. However, 555-5 does require that each single receptacle used for supplying shore power to boats be supplied by an individual branch circuit. By definition, an individual branch circuit could not have two receptacles.

Each receptacle (eight total, apparently) is required by this section to be supplied by its own circuit. Two or more receptacles could be connected to the same feeder, but each would have to have its own overcurrent device in order for each to be considered an individual branch circuit.

If, by circuit you were referring to a feeder, the feeder would have to be sized for both receptacles. Since the two receptacles are the same voltage, 555-6(a) does not seem to apply, so each receptacle would have to be counted in the load calculation for the feeder. However, if the two receptacles are intended to be used as in 555-6(a), that is, only one at a time, including only the larger load would seem reasonable. But that would not be in strict compliance with the language of the Code. Perhaps this is what you meant by referring to Section 555-6.
Noel Williams

Meet the Code Guys

Dann Strube is a nationally recognized NE Code expert and electrical code consultant.

Greg Bierals is president of the Electrical Design Institute, Davie, Fla.

Noel Williams has worked in electrical construction for more than 25 years, including 20 years experience in design/build projects. Licensed as an electrical inspector in Utah and a licensed master electrician in three states, he is co-author of the NFPA's 1999 NE Code Changes.