Arc flash study data collection: a field checklist

Every input IEEE 1584 needs, broken down by equipment type. Written by a licensed PE.

An arc flash study is only as accurate as the data going into it. I have watched experienced engineers do everything right in the office software, plug in one wrong percent impedance, and produce a study that tells someone to stand in a Category 4 zone when they should be in a Category 2. The calc itself is deterministic. The field data is where arc flash studies actually succeed or fail.

This post is the checklist I use for data collection on the walk. It is organized by equipment type because that is how you will move through the facility, and because IEEE 1584 needs a specific set of inputs per equipment class. Print it, or pull it up on your phone, or just read it once and internalize it. Every missed field here is a potential second site visit.

Start with the utility: the single most commonly missed input

The available fault current at the service point sets the ceiling for every downstream calculation. Underestimate it and every incident energy number in your study is low (unsafe). Overestimate it and you are specifying unnecessarily expensive PPE and equipment ratings.

The utility will not have this posted anywhere in the facility. You need to request it directly from the utility, using their standard short-circuit data request form. Collect:

• Three-phase available fault current at the service point, in amps
• Single-line-to-ground fault current at the service point
• X/R ratio for both fault scenarios
• Service voltage (primary side if you are behind the utility transformer, secondary if they own it)
• Minimum and maximum fault current estimates where the utility provides them, since arc flash calculations per IEEE 1584 typically run for both extremes

Request this 5 to 10 business days before your site visit. Some utilities respond same-day; others take two weeks and multiple phone calls. Do not let this be the critical path on your study schedule.

Main transformer

If the facility is fed through a utility-owned pad transformer and there is no customer-owned transformer, you can skip this section. Otherwise:

• Nameplate photo, full and clear
• kVA rating
• Primary voltage and secondary voltage
• Percent impedance (%Z), which is always on the nameplate but often in small print
• Winding configuration (delta primary, wye secondary is common; confirm)
• Tap setting if adjustable
• Approximate conductor length from transformer secondary to the main distribution bus

Main switchgear and main distribution

• Switchgear or panel nameplate photo
• Bus rating (e.g., 2000A)
• Bus voltage and number of phases
• Main breaker or main lug, with trip rating, frame size, and interrupting rating (AIC)
• If the main is electronic-trip, capture the trip unit type (LSI, LSIG, etc.) and the trip settings (long time pickup, long time delay, short time pickup, short time delay, instantaneous pickup)
• Every branch position on the bus: breaker trip rating, frame size, AIC, number of poles, and connected downstream equipment name
• Feeder conductor to each downstream bus: size, material (copper or aluminum), conduit type, and run length

The trip unit settings matter for arc flash. A breaker set to clear a fault in 0.05 seconds produces a dramatically different incident energy than the same breaker set to clear in 0.5 seconds. If the setting dials are accessible, photograph them. If the panel door must stay closed, ask the facility's O&M records for the commissioning or maintenance report that captured them.

Sub-distribution panels and MCCs

Same capture as main switchgear for each sub-bus:

• Nameplate photo plus panel schedule photo if one is posted inside the door
• Bus rating, voltage, phases
• Main breaker or MLO details
• Every branch position, trip rating, AIC, poles, connected load
• Incoming feeder conductor size and length from the upstream bus

For MCCs specifically, each bucket needs motor nameplate data (see next section) because motors contribute to downstream fault current and IEEE 1584 requires that contribution for a full calculation.

Motors (because they contribute to fault current)

Every motor larger than about 50 HP contributes meaningfully to the short-circuit current at the bus it is connected to, for the first few cycles after a fault. Small motors do too in aggregate. For an accurate arc flash calculation, you need:

• Motor nameplate photo
• Horsepower and full load amps (FLA)
• Voltage and phase
• Locked rotor code (NEMA letter) or locked rotor current if listed
• Service factor
• Subtransient reactance (X''d) if listed (rare on standard induction motors but common on large synchronous and specialty motors)
• Associated starter (across-the-line, VFD, soft-start) and its location

Industry practice is to group small motors (under 50 HP) by voltage and use aggregate contribution values. Larger motors get modeled individually. Capture everything 50 HP and up by nameplate; capture smaller motors as counts and aggregate HP per bus.

Generators, UPS, and alternate sources

Emergency power sources change the fault current picture when they are online. You need the data for both configurations (utility-only and generator-backed) because arc flash values can differ significantly.

• Generator nameplate photo
• kW and kVA rating
• Voltage, phase
• Subtransient reactance (X''d), transient reactance (X'd), and synchronous reactance (Xd) if listed
• Power factor
• Fuel type, tank size, runtime at full load
• Which loads run on the generator (every transfer switch it feeds, clearly named)

Transfer switches

• Nameplate photo
• Rating (amp rating of the switch itself)
• Number of poles
• Type: contactor, breaker-based, or closed-transition
• Withstand and closing rating (this is the arc flash and short-circuit rating for the switch body itself)
• Normal source feeding the switch (panel or bus name, feeder size, length)
• Emergency source feeding the switch (generator name, feeder size, length)
• Downstream load (what the switch feeds)

Conductors, raceways, and cable runs

Cable impedance contributes to the fault current drop between buses. For a rigorous study, every feeder between two buses needs:

• Conductor size and count (e.g., 3 sets of 500 kcmil copper)
• Material (copper or aluminum)
• Conduit type (steel, PVC, aluminum, EMT) since it affects reactance
• Approximate run length in feet
• Number of conductors per phase (parallel sets)

Measure rather than guess. A 50-foot error on a 3-set 500 kcmil run through steel conduit will shift the downstream fault current by a meaningful amount.

The five most common second-visit triggers

1. Percent impedance missing from a transformer nameplate photo. Small print, people zoom past it.
2. Interrupting rating (AIC) missing on branch breakers. Trip rating is easy; AIC takes a closer photo.
3. Feeder lengths guessed instead of measured. Pace it or use a laser.
4. Trip unit settings not captured on electronic-trip breakers. If you cannot read the dials with the door closed, get the commissioning report.
5. No utility short-circuit data. If you walked the site without having already requested it, you are not done.

What a clean field package looks like at the end of the walk

If you capture everything above, your field package ready for the office is:

• A structured equipment list (spreadsheet or tool) with every nameplate field
• Photos of every nameplate, tagged with the equipment name
• Utility short-circuit data letter from the utility
• Feeder conductor sizes and measured lengths for every cable run on the one-line
• Trip unit settings for every electronic-trip breaker
• A rough connectivity map (or full one-line) of the distribution system

With that in hand, you can build the IEEE 1584 model at your desk without going back.

How I run this now

I built AmpSketch because the structured data capture above was always the limiting factor for me. You can have a perfect checklist and still lose time transcribing notes and matching photos to equipment three days later. The tool turns the walk into structured capture: open it on your phone at each piece of equipment, snap the nameplate, AI reads the values into the right fields, you verify, move on. The one-line builds as you walk.

A CSV equipment list export for hand-off to SKM, ETAP, or EasyPower is on the roadmap, along with a more complete on-device arc flash workflow. Today the tool collects the data and builds the one-line; the arc flash calc itself still runs in your existing study software. The integration story gets cleaner as we ship the next few releases.