By Jay Prigmore II, Ph.D., P.E. Last reviewed · Prepared and reviewed by a licensed Professional Engineer
Incident energy analysis, arc flash boundaries, and PPE categorization, performed and reviewed to IEEE 1584 and NFPA 70E by the engineer described below.
By Jay Prigmore II, Ph.D., P.E. Last reviewed · Prepared and reviewed by a licensed Professional Engineer
IEEE 1584, the Guide for Performing Arc-Flash Hazard Calculations, is the method most widely used to calculate incident energy and arc flash boundaries. PEFG's principal, Jay Prigmore, Ph.D., P.E., is a voting member of the working group that develops and revises it, and a founding member of the NFPA Technical Committee on Electrical Inspection Practices.
That matters most where the standard is ambiguous, where the 2018 and 2002 editions diverge, or where a system falls outside the parameter ranges the empirical model was derived from. Those are the situations where a study is most likely to be wrong and least likely to be caught.
In review work and in litigation, the same defects recur:
Where an arc flash injury has occurred, the technical questions are usually whether the incident energy was correctly calculated, whether the PPE specified was adequate for the actual exposure, whether the labeling reflected the as-installed system, and whether the employer's electrical safety program met NFPA 70E. PEFG performs incident energy reconstruction for these matters and provides deposition and trial testimony (see expert witness services).
PEFG publishes free arc flash and power system calculators at powerengcalc.com, including IEEE 1584-2018 and 2002 incident energy, DC arc flash, and electrode configuration references. Those tools are for engineering estimation and education. A stamped study for labeling and compliance requires a full system model and a licensed engineer's review.
Incident energy is not a property of the equipment. It is a property of the equipment, the available fault current, the arcing current that fault current produces, the clearing time that arcing current causes, the electrode configuration, the enclosure dimensions, and the working distance, and several of those are coupled.
The coupling is what produces disputes. Arcing current is lower than bolted fault current, and a lower current sits further out on the protective device’s time-current curve, which means a longer clearing time, which means more incident energy. IEEE 1584 addresses this with a second calculation at a reduced arcing current, and the higher incident energy of the two cases is used. A study that skipped the reduced-current case will read low, and it will read low specifically in the range where a device is about to fall off its instantaneous element, which is where an understated result matters most.
The 2018 edition changed the underlying model materially: electrode configuration became an explicit input, the enclosure size correction was reworked, and the valid parameter ranges shifted. A 2002-edition study is not wrong for having been performed under the 2002 edition. It becomes a problem when it is relied on today as though the numbers were still current, or when it is compared against a 2018-edition result as if the two were computed the same way.
The incident-energy arithmetic is public. The IEEE 1584 arc flash calculator at PowerEngCalc shows the full derivation alongside the result, with a side-by-side 2018-versus-2002 edition comparison and a breakdown of electrode configuration effects. That input most often explains a disagreement between two studies. For DC systems, the DC arc flash calculator and the note on choosing a DC method cover ground IEEE 1584 does not.
If you are holding a study of unknown quality and need a triage before committing to a full expert review, Ohmics runs an automated review that flags edition, missing reduced-current cases, out-of-range parameters, and label inconsistencies at a fraction of the cost. It is a first pass and not an expert opinion. It can show quickly whether a full review is warranted.
Questions that come up before a retention decision, answered by the engineer who would do the work.
The 2018 edition replaced the empirical model at the core of the guide. Electrode configuration became an explicit input: vertical conductors in a box, horizontal conductors in a box, vertical conductors terminating in a barrier, and the open-air equivalents each produce materially different incident energy for otherwise identical conditions. The enclosure size correction was reworked, the valid ranges for voltage, gap, and fault current shifted, and the calculation now spans a wider system range. The practical consequence is that a 2002-edition result and a 2018-edition result for the same equipment are not directly comparable, and treating them as though they were is a common analytical error in litigation.
NFPA 70E requires the incident energy analysis to be reviewed at intervals not to exceed five years, and whenever a major modification or renovation takes place. In practice the five-year clock is the less important trigger. Any change to the available utility fault current, a service or transformer replacement, a change to protective device settings, or the addition of on-site generation can change incident energy immediately, and none of those events announce themselves to whoever holds the study. A study that is three years old and predates a transformer swap is more out of date than one that is five years old and unchanged.
Yes, and often with better inputs than the pre-incident study had, because the as-found system configuration is known rather than assumed. The reconstruction uses the actual protective device settings as found, the actual clearing time from the relay or trip unit record where one exists, the actual working distance and body position derived from the injury pattern and witness accounts, and the actual electrode configuration of the equipment involved. The result can then be compared with the injuries and the condition and arc rating of the clothing worn. That comparison is approximate, because injury severity depends on more than incident energy, but it is a check the original study did not have.
No. A label is one output of a compliance program, not the program. NFPA 70E requires a documented risk assessment, an electrical safety program, training appropriate to the task, correct PPE selection and condition, an energized work permit process where energized work is justified, and equipment maintained so that protective devices actually clear in the time the study assumed. A facility can be fully labeled and still fall well short, most commonly where the labels reflect a study whose assumed device settings were never actually applied in the field.
Almost always because of inputs rather than arithmetic. The usual sources are a different IEEE 1584 edition, a different available utility fault current, protective device settings modeled as designed rather than as found, electrode configuration defaulted rather than determined per location, a different working distance, and whether the reduced-arcing-current case was run. Reconciling two studies is generally a matter of putting the two input sets side by side; the equations themselves are rarely where the disagreement lives.
Only within the conditions it was tested under. IEEE C37.20.7 arc-resistant construction is qualified for specific accessibility types, specific arc durations, and only with all doors and covers closed and latched as tested. Equipment operated with a door open, with a compartment cover removed, or beyond the tested arc duration is not being used in the configuration that was qualified. Whether the equipment was in its qualified configuration at the moment of the incident is usually a factual question the physical evidence can answer.
New studies, updates to stale studies, and independent review of work performed by others.