Run two calculations, not one. Model the arc on the line side of the main device with the main cubicle or incoming enclosure, and the arc on the main bus downstream of it with the smallest compartment connected to that bus. A metering or TVSS cubicle is usually the smallest. IEEE 1584-2018 asks for the size of the box around the arc: the equations are normalized to a typical enclosure, the enclosure size correction factor is 1.00 at that size and below, and it falls as the opening grows toward the standard's cap. The compartment you enter, not the lineup's outer steel, decides the number on the door. Here is how to pick the size, run both cases, and label them.
What enclosure size does IEEE 1584-2018 ask for?
It asks for the equivalent width and height of the enclosure opening at the arc, entered in inches against a typical box of 20 x 20 x 20 in [508 x 508 x 508 mm], with both width and height capped at 49 in [1245 mm]. The 2002 edition assumed a single box size, so size never reached your input screen. The 2018 edition was fitted to tests across a range of box sizes and added the enclosure size correction factor: a multiplier on calculated incident energy that you feed with the opening's width and height. Your software pulls its defaults from the standard's table of typical bus gaps and enclosure sizes by voltage class — 32 mm [1.26 in] is the typical gap for low-voltage gear — and you override the box with the real compartment.
| What you measure at the opening | What goes into the model |
|---|---|
| Rectangular compartment | Its actual width and height |
| Width or height above the cap | Entered at the cap |
| Compartment at or below the typical box | Real dimensions; the factor holds at 1.00 |
| Non-rectangular opening | Equivalent width and height: the average of four measurements on each axis |
All three enclosed electrode configurations take the same size entry: vertical conductors in a box (VCB), vertical conductors ending in an insulating barrier (VCBB) and horizontal conductors in a box (HCB). The configuration you choose moves the answer more than the size does — a published worked comparison of one fault returns 105.0 cal/cm2 for VCBB and 145.3 cal/cm2 for HCB, a 38 % difference from configuration alone. Keep these entries on the arc-physics side of your desk; the enclosure sizing guide handles the bending-space and fill side.
Which size do you enter when the equipment has compartments?
Enter the compartment the arc sits in: main cubicle dimensions for the line-side case, and the smallest connected cubicle for the downstream bus case, so a lineup gets two calculations rather than one. An MCC, switchboard or switchgear lineup is a row of separate boxes sharing one bus, and the model's question — how big is the box around this arc — has a different answer at every door. The pattern that fits the equipment you actually meet:
| Arc you model | Enclosure size you enter | Where the result lands |
|---|---|---|
| Line side of the main breaker or switch | The main cubicle or incoming enclosure; if the incoming section is one open volume behind one door, that volume's outer dimensions | Label on the main cubicle door |
| Main bus downstream of the main | The smallest cubicle connected to that bus, often a metering or TVSS compartment | Label on the downstream doors of the lineup |
The same logic scales down to a small box: arc flash with several circuits in one junction box runs the equivalent decision where the compartments are terminal barriers instead of cubicles.
Why does the smallest cubicle govern the downstream label?
Because the factor sits at its maximum, 1.00, in any compartment the size of the typical box or smaller, the tightest cubicle always returns the highest calculated energy for the same fault. Calculated energy is proportional to the factor, so your comparison of two compartments is a ratio of their factors — the worked example below runs that division. The physics behind it: a small opening keeps the plasma and radiant heat aimed at whoever stands at the door, while a wide opening lets the energy spread and cool. Downstream of the main device, any door on the bus can be opened for troubleshooting or thermography, so the bus label carries the worst compartment on that bus. That two-case pattern — main cubicle for the line side, smallest cubicle for the bus — is the settled field practice on this equipment.
Do breakers and partitions inside the compartment count?
Enter the compartment as an open box: the 2018 equations were fitted to tests on empty enclosures, so the method gives you no input that credits barriers, breakers or contactors with absorbing any of the arc energy. Real compartments are full of steel that would reflect and absorb something, but the published equations have no term for it, and a hand reduction will not survive a study review — do not deduct anything for that steel.
One judgement call belongs to the line-side case: a single open volume behind one door takes the volume's outer dimensions, and a barrier that closes the incoming section into its own cubicle makes the cubicle your box. Incoming line compartment design covers building that section so the study and the steel agree. Keep the concepts separate, too: the size factor is a calculation input, not a claim about surviving an internal arc — what an arc-rated enclosure is covers that different question.
How do you run the two-case calculation?
Six steps: mark the two arc locations, measure each opening, apply the entry rules, run both cases, compare the energies, then issue two labels where the results differ.
- Mark the two arcs on your one-line: line side of the main device, and the main bus downstream of it.
- Pull each compartment's opening dimensions from the enclosure drawing — width and height in inches; for a non-rectangular opening, average the four measurements on each axis.
- Apply the entry rules: cap any width or height above the 49 in [1245 mm] limit, and let the factor sit at 1.00 in compartments the size of the typical box or smaller.
- Run both cases in your software, each on its own clearing path — the upstream device clears the line-side arc, the main device clears the bus arc — with only the box dimensions changing between the cases.
- Compare the two incident energies at the same working distance; the higher number governs the label where it applies.
- Record the modeled location, the dimensions entered and the no-credit decision for obstructions in the study file.
What does the calculation look like with real numbers?
A 480 V switchboard, 22 kA bolted fault, VCBB electrodes at the standard's typical 32 mm [1.26 in] gap, 0.08 s clearing, 18 in [457 mm] working distance: 6.2 cal/cm2 on the 36 x 49 in main-cubicle entry and 7.95 cal/cm2 in a 14 x 16 in TVSS compartment. Hold the fault current and clearing time equal in both runs so the enclosure is the only thing moving.
- Main cubicle: 36 in wide x 96 in high. Height caps at the limit, so you enter 36 x 49 in; the software returns a factor of 0.78 and incident energy of 6.2 cal/cm2 [26 J/cm2].
- Smallest compartment on that bus: a 14 x 16 in TVSS cubicle, below the typical box, so its factor is 1.00.
- Small-compartment energy: 6.2 x (1.00 / 0.78) = 6.2 x 1.282 = 7.95 cal/cm2 [33 J/cm2].
- Labels: 6.2 cal/cm2 line-side on the main cubicle door, 7.95 cal/cm2 on the downstream doors, each with its arc flash boundary.
The 1.75 cal/cm2 gap is the enclosure alone, a 28 % rise. Start the main-cubicle run at 6.3 cal/cm2 and the small compartment returns 8.1 cal/cm2 — across the 8 cal/cm2 minimum arc rating of NFPA 70E's PPE category 2 — so the compartment you model decides what the electrician wears.
What goes on each arc flash label?
Two labels when the numbers differ: the main cubicle door carries the line-side incident energy, and every door on the bus downstream carries the small-cubicle result, each with its arc flash boundary, as NFPA 70E requires for equipment worked on energized. NFPA 70E asks the label to carry the incident energy at the working distance or the PPE category, plus the boundary and the nominal voltage. Before the labels go on, check four things:
- The dimensions you entered match the enclosure drawing — each compartment's opening width and height, with the cap applied.
- Both cases appear in the report: line side of the main, and the smallest cubicle downstream.
- A single label covers the lineup only when both cases return the same energy; otherwise the two labels differ.
- Each label names the working distance its incident energy was calculated at.
Three mistakes undo these runs. Entering the lineup's outer dimensions everywhere understates every small compartment by the factor ratio, 28 % in the example. Entering the full 96 in height instead of the capped height leaves the model's tested range. And a hand reduction for internal steel has no basis in the published equations. When the study sends you back to the steel — a compartment that must grow, a barrier that must move — custom enclosures built to your dimensions from one piece, with every cutout, are delivered in 10 business days counted from your approval of the drawing. Before anyone opens a door, match the task to the label: assessing the work first for arc flash PPE decides what gets worn for that job, and taking the task de-energized whenever the schedule allows beats any label.
Catalog dimensions to compare
Compare these catalog dimensions with your component layout. Open each enclosure for its rating, material and usable space, then check those specifications against your application.
| Enclosure | Material | Outside H x W x D / display | Rating |
|---|---|---|---|
| 12 × 10 × 6 in | 304 stainless | 12 x 10 x 6 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 20 × 16 × 8 in | 304 stainless | 20 x 16 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 24 × 20 × 8 in | 304 stainless | 24 x 20 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
Questions
Does enclosure depth change the 2018 correction factor?
No. The factor is computed from the equivalent width and height of the opening, the average of four measurements on each axis when the opening is not rectangular. Depth shapes your result through the electrode configuration you select, VCB, VCBB or HCB, not through the size factor, and all three enclosed configurations use the same width and height entry rules.
What do you enter for a compartment wider or taller than 49 inches?
Enter 49 in [1245 mm] for that dimension. The tests behind the 2018 equations covered boxes only up to that cap, so a 36 in x 96 in main cubicle goes in as 36 in x 49 in and a 60 in wide section enters at the cap. The factor bottoms out at the cap, which is why very large lineups stop gaining.
Does the enclosure factor apply to equipment in open air?
No. IEEE 1584-2018 also defines open-air configurations, vertical or horizontal electrodes in open air, and they carry no enclosure size correction factor because there is no box to concentrate the energy. The factor belongs to the enclosed configurations only. Open bus and open equipment are modeled with their own coefficients at the typical working distance for that class.
How much can the compartment choice change the incident energy?
Divide one factor by the other: calculated energy is proportional to the factor, so a move from 0.78 to 1.00 multiplies your result by 1 / 0.78 = 1.28, turning a 6.2 cal/cm2 cubicle result into 7.95 cal/cm2 in the small compartment. Run the check in reverse and a study that entered the lineup's outer dimensions everywhere understates every small-compartment door by that same 28 %.
