Arc Flash, Coordination and Short Circuit Study for a Federal Campus, Houston (2019, Updated 2024)

In 2019 we completed an arc flash, coordination and short circuit study for a multi-building federal campus in Houston, Texas, and in 2024 we updated it. Client details are confidential.

One campus, six ways to run it

The campus houses equipment, security, offices and leased space. Two utility feeds, each through a 3,000 kVA transformer, deliver 480Y/277V service to the main switchgear, and two 2,000 kVA standby generators back the site up without running in parallel with the utility. Tie breakers, automatic transfer switches and a UPS bypass mean the same bus can be fed in several different ways, and each arrangement gives a different fault current. We modeled six operating configurations so that every bus was studied under the source arrangement that is worst for it. A manual transfer switch was set to its most conservative source in every case.

Building the model

System data came from the owner one-line diagrams, from walkdowns by our engineers together with facility personnel, and from fault data supplied by the serving utility. Every major bus was modeled in ETAP with the equipment around it, including cable lengths and sizes, transformers, panels, automatic transfer switches and generators. Where a breaker could not be reached without an outage, we modeled it with no trip function, which gives a conservative answer. Typical values stood in for transformer X/R ratios and for motor and generator impedances, and typical enclosure and bus gap dimensions from IEEE 1584-2018 were used instead of measurements on energized gear.

What the study calculated

We calculated three-phase and single-line-to-ground fault currents at each bus. Equipment duty was checked against ANSI C37.010, and low-voltage breaker interrupting ratings against C37.13 and UL 489. An electrical engineer set the coordination breaker by breaker, and a settings document lists every trip and relay setting the coordination depends on. Incident energy and arc flash boundaries followed IEEE 1584-2018 and NFPA 70E-2018, with the arcing time capped at 2 seconds, as IEEE 1584-2018 allows where a worker can move away from the arc.

Findings the owner had to act on

One device needed its interrupting rating verified against the available fault current before the labels went up. The primary utility and standby generator buses came out above 40 cal/cm², and no breaker setting change could bring them down with the equipment installed. For those buses we recommended arc flash protection added at a future upgrade, either a light-sensing system that trips a shunt-trip breaker or relays with arcing fault detection. We also asked the owner to photograph the panels whose breaker data could not be collected without an outage, so the model could replace conservative assumptions with real settings.

The five-year update

NFPA 70E calls for the incident energy analysis to be reviewed at least every five years and whenever the system changes. In 2024 we updated the model, the reports and the label files. The update lowered the settings on the two main breakers: ground fault pickup from 0.6 to 0.4, short-time pickup from 4 to 3 and long-time delay from 10 to 3.5. Coordination with the utility improved, but clearing at those buses was still slow, so the update repeated the case for dedicated arc fault sensing. The 2024 deliverable again covered six operating scenarios.

Keeping the labels true

A label is only as good as the model behind it. What a study includes is set out on the page for our arc flash study services. The settings side is covered under our protection and coordination work.