Topics discussed in our "Code Forum" column involve complicated issues requiring extensive analysis. However, not every Code question warrants such in-depth treatment. Here are the latest short answers to questions posed on our website. Coverage includes topics in: Sec. 220-35, 250-96, 250-97, 305-4(c), 422-11(e)(3), 630-12.

Q. We are an engineering/architectural firm. One of our projects involves remodeling supermarkets. The scope of the design includes replacing refrigerators with new units, upgrading the lighting, and revising the produce, bakery, and checkout areas.

Can we apply NEC Sec. 220-35 "Optional calculations for additional loads to existing facilities" where it is permitted to use the past one-year historical maximum demand obtained from the utility for determining the adequacy of the existing service?

Up to now, we have been applying Sec. 220-35 by obtaining the maximum 12-month historical demand data times 125% and adding the new loads to determine if the electrical service to the store is adequate. Is this method correct?

Since several of these supermarkets have already been completed, we obtained the utility demand (past three to four months' worth) and found it to be running the same percentage as before the remodeling.

A. Your work is exactly what this section is intended for, and it appears you're applying it correctly. It's always nice to see a Code rule applied correctly, but good follow-up data that validates the procedure makes it even better.

You're fortunate to have the full year's data available. For example, until the 1996 NEC increased the use of this section substantially, it's been much more difficult for designers looking for a similar process on a feeder without separate metering. If the full year isn't available, this section now (in an exception) allows a 30-day analysis. If you do this, be sure to add any loads that are out of season to your calculation.

Q. My question concerns using 2-in. EMT from a busplug in our factory to a disconnect switch supplying a 480V receptacle. If I pull a ground from the source to the disconnect switch, must I use grounding bushings, even though I pulled a separate ground?

Also, do you have to use a bonding bushing when installing steel flex with an equipment ground to a transformer?

A. It depends on Sec. 250-96 and Sec. 250-97. If the voltage isn't more than 250V to ground and the connector-to-enclosure ground path is secure, then you do not have to install the bonding bushing. The same rule goes if the voltage is higher and the connector is solidly grounded by a locknut at a clean knockout or to a multiple knockout in an enclosure listed as having suitable grounding continuity through the concentric knockouts.

Note: These rules apply whether or not there is a contained equipment grounding conductor, because the armor has to be grounded. And under fault conditions, it may see substantial current due to skin effect, even though it has substantial impedance. Sec. 250-96(a) also requires this grounding (secure grounding to the conductive wiring method), whether or not you use supplementary equipment grounding conductors.

Q. I have a question about wiring GFCI devices with grounded and ungrounded, single-phase power supply systems. I want to install a GFCI in my industrial machine, remote from the power service panel. The machine requires 240VAC single-phase, and the frame isbonded to earth ground. Will the GFCI work properly if I tie the pigtail to earth ground instead of a neutral line?

A. If you mean a 120V GFCI-protected circuit originating at a GFCI circuit breaker (the only such device with a "pigtail"), the answer is no; for many reasons unrelated to the functioning of the GFCI. You would be using the equipment grounding conductor to the machine as the neutral (technically not a neutral here) return from the GFCI circuit, which is one of the biggest "no-nos" in grounding. That's because the return current would elevate the voltage of that grounding system (by the product of current, voltage, and power factor) as it passed over the machine equipment ground ahead of the breaker. That potential shock hazard over time on the supply wiring method would be invisible to the GFCI device on its load end.

You could bring a neutral with the motor circuit conductors, but you would have to evaluate the circuit load to see if the branch-circuit conductors to the motor could additionally function as feeder conductors to this circuit breaker. If you reground in this manner, you will almost certainly violate some Code rule. You should either arrange a conventional feeder to the machine, or bring a stand-alone branch circuit over with the machine circuit. Keep in mind this might raise the question of more than three current-carrying conductors in the same raceway.

Q. We are renovating a small dental office. For temporary lighting, we are using UL-approved plastic sockets with plastic lamp guards. The socket works by unscrewing the top and running THWN or similar wire through the socket. By screwing down the top, the socket makes the connection without cutting the wire. We are suspending the wire by the socket. Is this approved by the NEC?

A. No; not since Sec. 305-4(c) was revised (1996 NEC) to eliminate individual conductors for temporary branch circuits. Now, they have to be in a cable assembly. A construction worker in Pennsylvania was electrocuted due to individual conductors on a lighting string, as you apparently described, contacting the sharp edges of metal studs.

Q. If you have a single-phase welder 230V 100% duty cycle, with nameplate rating of 50A, what's the breaker and wire size? I would say a 50A breaker and No. 6 wire. I remember a piece of equipment that came into my plant to wire up. Let's say it had a couple of motors, some heat loads, solenoids, and other miscellaneous loads. It came with paperwork saying it was rated for 60A. Should I have increased that load by 125% to get my breaker size, could I have used the 60A for my breaker size and increase the 60A by 125% to get my wire size, or am I totally confused?

I ask because I had an argument with a coworker about a hand dryer in a rest room with a nameplate of 120V and 20A. I said increasing the amps by 125% would give you a breaker size of 25A or 30A, and the wire would be No. 10. He said he'd put the dryer on a 20A breaker and No. 12 wire. Who's right?

A. The welder has special rules in Art. 630: Sec. 630-12 would allow a 100A breaker ahead of wire that could be as small as No. 8, if the terminations were rated 75DegrC. Otherwise, you'd use No. 6.

The next question depends on whether or not you're looking at continuous loading. If so, the breaker goes up 125%, and the wire usually does the same. (This can get very complicated. For a complete analysis, refer to the August 1996 issue of EC&M.)

The Code-minimum answer to the last question (hand dryer) is No. 12, 20A; again because this isn't a continuous load. Of course, you can increase the wire size. Increasing the circuit breaker depends on listing restrictions (that may limit the protection to 20A) and Sec. 422-11(e)(3), which would create an absolute upper limit of 30A.