Tiza Group

Home / Guides / Codes and grid connection: US, UK, CA, AU

Arc Flash Study Data Collection Checklist

Arc flash study data collection checklist: what to gather on the one-line, sources, transformers, cables, protection and equipment before the study starts.

By Shadai García, CEO and protection engineer at Tiza Group.

infinite utility bus11,000 kVA, 480 VZ = 5.75%2480 V bus420.9 kA availablefeederrelay31 utility letter, 2 nameplates,3 relay read-out, 4 field photosPowerMeasure / tripfull load 1,203 A (1,000 kVA, 480 V)5.75%: 1,203 / 0.0575 = 20,918 A5.00% typed: 1,203 / 0.05 = 24,056 A24,056 / 20,918 = 1.15: +15% for one digitthe error moves arcing current and timeA010 kA20 kAfull load1,203 AZ = 5.75% (nameplate)20,918 AZ = 5.00% typed by mistake: 24,056 A, +15%utility treated as an infinite busData sourcesUtility letterNameplatesRelay read-outField photosStudy© tizagrp.cominfinite utility bus11,000 kVA, 480 VZ = 5.75%2480 V bus420.9 kA availablefeederrelay31 utility letter, 2 nameplates,3 relay read-out, 4 field photosPowerMeasure / tripfull load 1,203 A (1,000 kVA, 480 V)5.75%: 1,203 / 0.0575 = 20,918 A5.00% typed: 1,203 / 0.05 = 24,056 A24,056 / 20,918 = 1.15: +15% for one digitthe error moves arcing current and timeA010 kA20 kAfull load1,203 AZ = 5.75% (nameplate)20,918 AZ = 5.00% typed by mistake: 24,056 A, +15%utility treated as an infinite busData sourcesUtility letterNameplatesRelay read-outField photosStudy© tizagrp.com
Arc flash data collection: a 1,000 kVA 480 V transformer at 5.75 percent with its feeder breaker, CT and relay, the four data sources marked at the equipment, and the available fault current of 20,918 A against 24,056 A when the impedance is typed as 5.00 percent. Location in the system and operating principle. Illustrative: real values come from the nameplate and the manufacturer's manual.
Short answer

An arc flash study is only as good as its data. Before the calculation starts, someone has to confirm the one-line diagram against the plant, collect the utility fault data, transformer and cable data, the real settings of every protective device and the physical details of each piece of equipment. The data falls into six groups, each with a usual source, and a single wrong nameplate entry can shift the available current.

An arc flash study starts long before the calculation. Missing or doubtful data delays the study and can put a wrong number on a label. The arc flash study data collection checklist below groups what to gather and where it normally comes from. It also says what to confirm in the field, so the engineer who runs the model starts with a clean file.

The study itself follows the method in the IEEE 1584-2018 calculation guide. The rules that make the study necessary are in NFPA 70E 2024: what engineers need to know. For the wider framework, start at the US electrical codes and grid interconnection guide.

Short answer

The data an arc flash study needs falls into six groups: a verified one-line diagram, the utility source data, transformer and generator data, cable and bus data, protective device models with their loaded settings, and the equipment details at each study location. Verify the one-line against the plant first, because every other item hangs from it. Settings must come from the device, not from a drawing.

1. The one-line diagram, verified

The one-line diagram is the map of the study. Compare it with the plant:

  • Every source: utility service, generators, solar or storage with inverters, and which can run in parallel.
  • Every switching point: breakers, ties, transfer switches, normally open points and their real normal state.
  • Every bus and equipment line-up that will be a study location.
  • Voltage levels and the transformer connections between them.

A walk-down with photos of nameplates is the reliable method. When it is not possible, record which parts of the diagram were taken from drawings only. The guide on reading a single-line diagram helps when the plant has no up-to-date one.

2. The utility source

Ask the utility, in writing, for the available fault current at the service point, for the maximum and the minimum case, and for the ratio of reactance to resistance (X/R). Record the date and the system conditions behind the values. The minimum case matters, because lower current can mean slower clearing. The guide on short-circuit currents for protection settings explains the two cases.

3. Transformers, generators and large motors

  • Transformers: kVA, primary and secondary voltage, connection, percent impedance and, if known, X/R.
  • Generators: kVA, voltage, subtransient reactance and how they are grounded.
  • Motors: horsepower or kVA and voltage for motors large enough to feed a fault.

Take the numbers from the nameplate or the test report. Typical values are a fallback, and the report must say so. The guide on equipment data for a short-circuit study lists the fields.

4. Cables and busbars

Length, conductor size, material, number per phase and the type of raceway. Lengths scaled from a drawing are acceptable, if the tolerance is noted. Busbars and busway need rating and length.

5. Protective devices and their settings

Check this group with the most care, because a drawing can show an older setting. For every device that can clear a fault in the study:

  • Manufacturer, model and catalog number.
  • For a molded-case or power breaker with a trip unit: frame, sensor and plug rating, and every setting dial or menu value, including instantaneous and short-time.
  • For a relay: the function enabled, the pickup, the curve and time dial, and the instantaneous setting.
  • For a fuse: class, ampere rating and manufacturer.
  • Instrument transformer ratios that feed the relay.
  • Any arc-energy reduction setting, such as a maintenance mode switch, and its state.

The settings must be the ones loaded in the device. If nobody has the file, read them from the relay. The guide on relays without documented settings describes how. Keep a dated copy afterwards, as in IED settings backup and recovery.

6. Equipment details at each location

For every panel, switchgear cell, motor control center or other equipment:

  • Voltage and the equipment type.
  • Enclosure height, width and depth.
  • Conductor gap and the arrangement of the conductors (vertical or horizontal), when it can be determined.
  • Working distance for the tasks done there.
  • Whether the work is done with doors open, or only with devices operated closed.

Photos of the open equipment, taken by someone qualified, save a second visit.

Worked example: one nameplate value

This example uses illustrative data, not from any project. A 1,000 kVA, 480 V transformer has a nameplate impedance of 5.75%. Assume the utility is a very strong source (infinite bus) to isolate the effect of one entry.

  1. Full-load current: 1,000 / (1.732 × 0.480) = 1,203 A.
  2. Available fault current at the secondary with Z = 5.75%: 1,203 / 0.0575 = 20,918 A, about 20.9 kA.
  3. If the data sheet is entered as 5.00% by mistake: 1,203 / 0.0500 = 24,056 A, about 24.1 kA.

A single wrong digit raises the available current by 15% (24,056 / 20,918 = 1.15). That current drives the arcing current and the clearing time read from the curve. If the fault current sits near a knee of the relay curve, a 15% error can move the clearing time and, with it, the incident energy. The error costs nothing to catch if the nameplate photo is in the file.

The real system also has utility impedance and cable impedance, which lower the current. The example only shows the sensitivity to one entry.

Checklist for the engineer

  • One-line diagram compared with the plant, with differences recorded.
  • Utility fault data in writing, with date, maximum and minimum, and X/R.
  • Nameplate photo for every transformer and generator.
  • Cable and busbar lengths and sizes for every feeder in the model.
  • Make, model and loaded settings for every protective device, with the date they were read.
  • Current transformer ratios that feed each relay.
  • Enclosure dimensions and working distance for every study location.
  • A list of the data taken from typical values, so the report can flag it.
  • A named person who signed off the data package before the calculation began.

Who signs and seals

The study is engineering work that, in the United States, is normally reviewed and stamped by a professional engineer licensed in the project’s state. Tiza delivers the data review and the study ready for your engineer of record to review and stamp. Tiza does not sign, seal or certify work in the United States.

How Tiza can help

If you want to start with the data, Tiza runs power-system studies remotely and delivers them in English, ready for your engineer of record to review and stamp. We review your one-line, nameplates and settings, and tell you what is missing before the model is built. Contact us with what you have and we will scope it with you.

References

  • IEEE 1584-2018, IEEE Guide for Performing Arc-Flash Hazard Calculations, for the inputs the model uses.
  • NFPA 70E, 2024 edition, section 130.5, for the arc flash risk assessment.

Figure: data sources and what each feeds into the study, from the utility letter and nameplates to the relay read-out and the field photos.

Frequently asked questions

What data does an arc flash study need?

A verified one-line diagram, the utility's fault current and X/R ratio at the service, nameplate data for transformers, generators and large motors, cable and bus lengths and sizes, make, model and loaded settings of every protective device, and the type, enclosure size and working distance of each piece of equipment.

Why must the one-line diagram be verified in the field?

Drawings drift from the plant. A tie breaker added, a feeder moved or a generator connected can change the available current and the device that clears the fault. A walk-down or a documented comparison catches those differences before they enter the model.

Do I need the relay settings or is the drawing enough?

You need the settings loaded in the device. A setting sheet or a drawing may show an older value. For trip units and relays, read the settings from the device or from a dated backup, and record the date and who read them.

WhatsApp