The IEEE 1584 calculation, explained.

What IEEE 1584 is

IEEE Std 1584, Guide for Performing Arc-Flash Hazard Calculations, is the calculation method used to quantify how much thermal energy an electrical arc would deliver to a worker standing at a defined distance. Its output — incident energy in calories per square centimetre — is the number that drives PPE selection under NFPA 70E and equivalent workplace-safety frameworks worldwide.

The 2018 edition replaced the original 2002 model and is built on a far larger laboratory test program — over 1,800 arc tests — which is why its equations respond to details the old model ignored, most importantly the physical orientation of the conductors where the arc occurs.

Where the model applies

Outside these ranges — DC systems, transmission-class voltages, single-phase — other methods apply, and a competent study report will say so explicitly rather than quietly extrapolating.

The calculation pipeline

Step 1 — Verify the single-line diagram

Everything downstream inherits the quality of the single-line diagram. Field verification confirms transformer nameplates, cable sizes and routed lengths, breaker and relay models, and — critically — the protection settings as they are today, not as designed. The most common cause of a wrong label is not the math; it's a breaker whose settings were changed in the field and never made it back into any document.

Step 2 — Short-circuit study

The system is modeled (ETAP, SKM, or similar) and the three-phase bolted fault current is calculated at every bus: utility source impedance, transformer impedances, cable impedances, motor contribution. This is the same study that validates equipment ratings, which is why arc flash work is usually packaged with a short-circuit and coordination study rather than bolted on afterwards.

Step 3 — Estimate the arcing current

A real arc is not a bolted connection — the arc itself has impedance, so the arcing current is lower than the bolted value, dramatically so at low voltage. IEEE 1584 provides empirical equations for arcing current as a function of bolted fault current, voltage, conductor gap, and electrode configuration. At 480 V the arcing current can be roughly half the bolted value; at medium voltage the reduction is much smaller.

Step 4 — Determine the clearing time

The arcing current is traced onto the upstream protective device's time–current curve to find how long the fault persists. Fuse, breaker trip unit, or relay plus breaker operating time — this is where protection engineering meets safety engineering, and it is the single most influential input to the final number.

Step 5 — Calculate incident energy

With arcing current, clearing time, gap, enclosure dimensions, electrode configuration, and working distance, the model computes incident energy at the worker's position, in cal/cm². The enclosure matters because a box focuses energy toward the opening — the same arc in open air delivers substantially less energy to a worker than in a motor control center bucket.

Step 6 — Repeat at reduced arcing current

Arcing current varies in practice, so the 2018 standard requires the calculation be repeated at a reduced arcing current (using its arc-current variation factor; the 2002 edition used a flat 85% for low voltage). The reason is subtle and important: a lower current can push the protective device out of its instantaneous region into a slower part of its curve. The case that produces the higher incident energy governs the label.

Step 7 — Arc flash boundary and labels

The arc flash boundary is the distance from the equipment at which incident energy falls to 1.2 cal/cm² — the onset threshold for a second-degree burn. Each equipment item then gets a label showing, per NFPA 70E: nominal voltage, arc flash boundary, and the incident energy at working distance (or a PPE category). The full study report documents every assumption behind every label.

Electrode configuration: the 2018 game-changer

The 2018 model requires you to classify how the conductors are arranged at each possible arc location, because orientation changes how the arc plasma is driven — sometimes directly toward the worker:

HCB and VCBB configurations can produce markedly higher incident energy than VCB at the same bus — horizontal electrodes eject plasma outward, toward the working position. A study that assumed VCB everywhere (a common shortcut) can understate the hazard on exactly the equipment where workers rack breakers.

Why clearing time dominates

Incident energy scales approximately linearly with time. Current enters the equations with less than proportional effect — but time is pure multiplication. A fault cleared in 5 cycles (0.083 s) at a given current produces a fraction of the energy of the same fault cleared in 2 seconds.

This is also why arc flash mitigation is mostly protection engineering, not PPE procurement:

• Settings optimization — tightening instantaneous pickups where coordination allows, often cutting calculated incident energy at the worst buses by half or more at zero hardware cost.
• Maintenance-mode switches — a temporary fast setting group engaged while workers are at the gear.
• Zone-selective interlocking and differential schemes — fast clearing without sacrificing coordination.
• Remote operation and racking — moving the worker outside the arc flash boundary entirely.

What changed from 1584-2002 to 1584-2018

Practical consequence: any study performed to the 2002 model is methodologically superseded. If your labels predate 2018 — or your facility has changed — the five-year review clock in NFPA 70E is not the only reason to refresh.

The data checklist

What we need from a client (or collect in the field) before the math can be trusted:

1. Current single-line diagram — verified, or flagged for verification
2. Utility available fault current and X/R at the point of supply
3. Transformer nameplates: kVA, impedance, connection
4. Cable sizes, types, and routed lengths for significant runs
5. Protective device schedule: make, model, and as-found settings
6. Motor schedules above ~37 kW / 50 hp for contribution modeling
7. Per-bus physicals: enclosure type and size, electrode configuration, working distance
8. Operating scenarios: normal, standby generation, tie-closed configurations — each can produce a different worst case

Items 1 and 5 cause almost all study delays and almost all wrong labels. Our arc flash study service starts with an SLD verification pass for exactly that reason.

Frequently asked questions