What Is Incident Energy?

The short definition

Incident energy is the thermal energy delivered at a given working distance from an electric arc, expressed in calories per square centimetre (cal/cm²).

An arc flash is a release of energy from an unintended fault current arcing through air — producing intense heat, light, a pressure wave and shrapnel. Incident energy measures only the thermal part of that, and only at one specified distance.

That narrowness is deliberate and worth holding onto. Incident energy is not a measure of how dangerous an arc flash is in total. It says nothing about the blast pressure, nothing about the molten metal, nothing about the sound or the light. It answers one question: how much heat lands on a square centimetre of a worker standing at a stated distance. That question happens to be the one that PPE selection turns on, which is why it dominates the conversation.

Why calories per square centimetre

The unit is doing something specific: it describes energy per unit of skin area, not total energy released. That is what determines whether a burn occurs, and it is why the figure is always tied to a distance — the same arc delivers far less energy per square centimetre to someone standing further back.

A useful physical anchor: roughly 1 cal/cm² is about the exposure you would get holding a fingertip over a cigarette lighter flame for a second. Now imagine that delivered to the whole front of your body at once, and you have the shape of what an 8 cal/cm² event does to unprotected skin.

There is a third number worth carrying, though it is a rule of thumb rather than a threshold in the standard: above roughly 40 cal/cm², the blast and pressure effects of the arc become severe enough that arc-rated clothing alone is no longer a satisfying answer. Equipment in that range is where facilities start looking hard at de-energising, at remote operation, or at engineering the energy down.

What actually drives the number

Two things, and neither is the equipment's nameplate:

• Available fault current — the largest current the system can deliver at that point during a short circuit. It is a property of the system at that location: source strength (utility supply, transformers, contributing motors) and the impedance of the path.
• How fast the protective device clears — the arc stops when the upstream device opens. Clearing time is often the dominant term, because energy accumulates for as long as the arc burns.

Several other inputs shape the result — the electrode configuration, whether the arc is inside an enclosure and how big that enclosure is, the gap between conductors, the system voltage, and whether the system is AC or DC. An arc inside a box reflects energy back toward the opening rather than radiating away in all directions, which is why enclosed equipment produces higher incident energy at the worker than an equivalent arc in open air.

But the two terms above are the ones that move most, and the ones a facility can actually change.

Why clearing time usually wins

Energy is power multiplied by time. Double the duration of an arc and, roughly speaking, you double the energy delivered. This produces a result people find counter-intuitive:

The practical consequence is that incident energy cannot be eyeballed from a one-line diagram. Two panels fed from the same bus, with the same fault current available, can carry very different numbers because of how their upstream devices are set. This is the single most common reason a study result surprises the people who commissioned it.

It is also why a relay settings change made during an unrelated project — by an engineer solving a coordination problem, with no arc flash question anywhere in scope — can quietly invalidate a wall of labels.

Working distance is part of the answer

An incident energy figure is meaningless without the working distance it was calculated at. "8 cal/cm²" is not a fact about a panel; "8 cal/cm² at 18 inches" is.

Energy falls off sharply with distance. Moving further from the arc source reduces the energy reaching the worker considerably, which is the whole basis of the arc flash boundary and a large part of why remote racking and remote operation are favoured wherever they exist.

Working distance in a study is the distance from the arc source to the worker's face and chest, not to their hands. That distinction matters: hands and forearms are routinely closer to the arc source than the working distance assumes, and the incident energy there is higher than the label reports. It is one reason hand protection is treated separately rather than being covered by the same number.

This is also why an arc flash label reports the available incident energy and the corresponding working distance together. Read one without the other and the number cannot be applied to anything.

What incident energy is not

• Not a property of the equipment. It belongs to the system and the settings, at a location, on the day of the study. Replace every component inside an enclosure and the number can be unchanged; change a relay setting three levels upstream and it moves.
• Not a function of voltage. A 480 V system with a slow-clearing upstream device can carry higher incident energy than a higher-voltage system that clears quickly. Voltage drives the shock hazard and the approach boundaries; it does not drive incident energy directly.
• Not the arc flash boundary. The boundary is a distance, derived from incident energy. The two are constantly confused because they appear side by side on the same label.
• Not the same as a PPE category. A category is a bracket assigned by a table. Incident energy is a calculated value in cal/cm². They come from different routes through the standard.
• Not permanent. See the section on what invalidates a label.

What the number is used for

Incident energy feeds two decisions:

• The arc flash boundary — the distance at which it falls to 1.2 cal/cm². The boundary tool works this the other way round, from your own figures.
• PPE selection — the arc rating of clothing and equipment must at least match the incident energy at the working distance.

On that second point: the arc rating printed on a garment is a tested performance value, and matching it to incident energy is a floor rather than a target. A 12 cal/cm² exposure needs at least a 12 cal/cm² system of protection, counting the whole ensemble rather than one layer. Layering matters, and so does what is worn underneath — meltable synthetics beneath arc-rated outerwear defeat the purpose of the outer layer entirely.

The two routes, and why a label carries one or the other

This is the incident-energy-analysis route. 2027 NFPA 70E's alternative is the table method, which assigns PPE categories instead of calculating a value. A facility uses one route or the other for a given task — never both — which is why an equipment label carries an incident energy value or a PPE category, and never both for the same equipment.

The categories carry minimum arc ratings of 4, 8, 25 and 40 cal/cm², which is why people reach for them as a translation table. Resist that. The table method's rows are valid only within stated parameters — a maximum available fault current, a maximum clearing time, and a minimum working distance — and outside those parameters the row does not apply at all. Reading a category off a table and treating it as an incident energy figure skips exactly the check that makes the table safe to use.

Which route a facility takes is a program decision, documented in the electrical safety program. The analysis route costs more up front and produces equipment-specific numbers. The table route is faster and coarser, and stops being available as soon as your system falls outside the tables' parameters.

How incident energy gets reduced

Almost every effective approach attacks the clearing time, because that is the term with the most leverage.

• Faster protective-device response. Lowering instantaneous pickup, or adding a device that responds faster in the arcing-current range, cuts duration directly. The constraint is coordination: too fast and the system nuisance-trips.
• A maintenance-mode setting. A switchable setting that makes the upstream device far more sensitive while someone is working, and returns it to normal afterwards. It trades selectivity for safety during the window it matters.
• Arc-detecting protection. Relays that sense light and current together and open the circuit in a few milliseconds rather than waiting out a time-current curve.
• Current-limiting devices, which cut the fault off within the first half-cycle.
• Distance. Remote racking, remote operation, and extension tooling move the worker back, which reduces the energy that reaches them without changing the arc at all.
• De-energising. Not a reduction technique so much as the elimination of the exposure, and still the first answer the standard expects to be considered.

What does not reduce incident energy: bigger conductors, a larger transformer, a higher-rated breaker, or better PPE. The first three can raise it. The last one changes what happens to the worker, not what the equipment produces.

Reading it off a label

An arc flash label produced from an incident-energy analysis normally carries the incident energy in cal/cm², the working distance it was calculated at, the arc flash boundary, the nominal system voltage, the shock approach boundaries, and the date of the analysis.

Two habits are worth building. First, always read the working distance alongside the energy figure — they are one piece of information, not two. Second, always read the date. A label with no date, or a date that predates known system changes, is reporting a study that may no longer describe the equipment in front of you.

Why it changes without anyone touching the panel

Incident energy is system- and setting-specific. It is not a fixed property of the equipment, and it can change when nothing inside the enclosure has been altered.

That is why the analysis is expected to be reviewed when the electrical system changes, not only on a fixed calendar. An arc flash label is a snapshot valid for the system as it was studied. Treating it as permanent is how facilities end up with labels that are confidently wrong.

Separately, the arc flash risk assessment behind those labels is reviewed for accuracy at intervals not exceeding five years. The five-year clock is a backstop, not a substitute for reviewing after a change.

The practical fix is not a bigger calendar reminder. It is one question wired into the existing change-management process: does this change affect available fault current, protective-device settings, or clearing times? If yes, the study and the labels it produced need a look. That question costs nothing and catches the cases a five-year cycle misses entirely.

Where this goes wrong in the field

• Treating one panel's number as typical of the building. Incident energy is location-specific. Neighbouring panels routinely differ by an order of magnitude.
• Selecting PPE from a category while the label reports cal/cm². The two routes are alternatives. Mixing them mid-task is how people end up under-protected while believing they followed the standard.
• Ignoring the working distance. Standing closer than the study assumed puts the worker in higher energy than the label reports.
• Assuming a low number means low risk. Incident energy describes the thermal hazard only. The shock hazard is governed separately by voltage and the approach boundaries, and a low-energy panel can still kill someone.
• Letting settings changes escape review. The engineer adjusting a relay for coordination is usually not the person who owns the arc flash program.
• Reading an undated label as current. Without a date, there is no way to know what system it describes.

Reference material on incident energy in 2027 NFPA 70E. Values on your equipment come from a study of your own system; consult the current edition of the standard and your facility's arc flash risk assessment. Back to the 2027 NFPA 70E guide