An arc flash incident energy study produces a number: calories per square centimetre at a stated working distance. That number drives PPE selection, equipment labeling, and a great deal of compliance activity. After an actual incident, however, investigators frequently encounter injuries and equipment damage inconsistent with the number on the label. That inconsistency is not a nuisance. It is usually the single most informative piece of evidence available.
What the calculation actually assumes
IEEE 1584 models incident energy from available bolted fault current, arc duration, working distance, electrode configuration, and enclosure geometry. Of these, arc duration is by far the most consequential and the least certain. Duration is not a property of the equipment; it is set by how quickly a protective device detects and clears the fault. The calculation therefore embeds an assumption that protection will operate as designed and as coordinated.
When that assumption holds, calculated and observed energy tend to agree. When it does not — a mis-set trip unit, a mis-coordinated upstream device, a breaker that had degraded in service, a relay whose settings were changed and never re-studied — arc duration can be several times the assumed value, and incident energy scales with it. A study that was accurate the day it was performed can become badly wrong after a single settings change.
The engineering literature reflects how sensitive these results are to modelling choices. Work presented at the IEEE Electrical Safety Workshop has examined incident energy and PPE ratings in practice (Rockwell, IEEE ESW 2015), while more recent work extends the analysis to DC arc flash in battery energy storage systems (IEEE ESW 2023) — an increasingly common configuration that the classic AC methodology was never written to cover.
Reading the physical evidence
Forensic reconstruction does not begin with the model. It begins with what the arc left behind. Copper vaporized by an arc deposits as fine spatter and oxide in patterns that differ materially from copper melted by an external fire. Enclosure deformation records the pressure rise. Arc erosion at the terminals indicates where the fault initiated and how it travelled. PPE and clothing carry a thermal record of their own, and burn distribution on the injured worker is often the most direct measure of energy actually delivered at the working distance.
Those observations are then compared against the model. Where they diverge, the investigation turns to the protection system: primary-injection testing of the breaker, examination of the trip unit, review of relay event records, and comparison of as-found settings against the settings the incident energy study assumed. It is common to find that the study was performed years earlier, that the system was modified afterward, and that no one re-ran the analysis.
Why this matters in litigation
Arc flash matters routinely involve several parties at once: the employer, a contractor performing energized work, the equipment manufacturer, and whoever performed or maintained the arc flash study. The question of whether the incident energy label was correct — and whether the protection scheme could actually deliver the clearing time the label assumed — frequently determines how liability is apportioned.
The corollary is a practical one. Because arc duration dominates the outcome, the protective device is often more important evidence than the equipment that faulted. It is also the item most likely to be replaced during restoration, which is why preservation instructions need to reach the site before repairs begin.
For the underlying failure mechanisms and the full examination protocol, see our specialization area on arc flash and arc fault failure analysis.