An arc flash study is not one calculation. It is a sequence, and each stage depends on the one before it. Understanding the sequence is the easiest way to tell a thorough study from a fast one, and to know what you are being quoted.
The six stages
1. Data collection
Everything downstream is built on this, and it is where most studies go wrong. What gets gathered: utility available fault current at the service point, transformer nameplate data and impedances, conductor sizes, types and lengths, protective device manufacturers, types, ratings and as-found settings, and the actual topology of the system rather than the drawn one.
On an existing facility this has to happen in the field. Drawings are routinely out of date, breakers have been replaced with whatever was available, and settings have been adjusted by people who are no longer there. On new construction there is nothing to walk, so the study is built from equipment submittals instead.
We send a licensed P.E. to collect this data, with an electrician as support to open panel covers. Missing conductor lengths and unrecorded breaker settings go straight into the incident energy result, and the engineer who has to defend the model is the one who should gather its inputs.
2. Single-line diagram
The collected data becomes a single-line diagram, either created from scratch or reconciled against what you already have. If your existing drawing disagrees with the field data, the field data wins and the drawing gets corrected. You keep the updated single-line, which for many facilities is worth as much as the study.
3. Short circuit study
Available fault current is calculated at every bus and compared against the interrupting rating of the equipment installed there. This is where underrated equipment surfaces — a breaker applied above its interrupting rating is a safety problem independent of arc flash, and it is cheaper to find on paper than during a fault. Also produces the maximum available fault current required by NEC 110.24 to be field marked on service equipment.
4. Coordination study
Protective devices are plotted on time current curves so that a fault clears at the device closest to it rather than tripping something upstream and taking out half the plant. Coordination also directly affects incident energy: how fast a device clears determines how long the arc burns, and duration drives the energy released.
Ask any firm whether every panel appears in the TCC plots or only the main gear. This is the stage most commonly cut short.
5. Incident energy analysis
Using IEEE 1584 methodology, incident energy in cal/cm² is calculated at each working location, along with the arc flash boundary and the PPE required. This is the output people think of as "the arc flash study," and it is entirely dependent on stages one through four being right.
6. Labels, report, and seal
Labels meeting NFPA 70E and NEC 110.16 are printed and installed on the correct equipment, with location-matched identifiers and the values calculated for that specific gear. The report is sealed by a California licensed Professional Engineer. You also receive the model itself, so future changes can be evaluated without rebuilding it from scratch.
How long it takes
New construction: a small project can be turned around in about one week. There is no site visit, because the study is built from equipment submittals — which is also why sequencing matters. Where Division 26 05 73 is in the specifications, the study is required before equipment is ordered.
Existing facilities: longer, and honestly variable. The schedule is driven by field data collection, which has to be planned around production, and by how much of your existing documentation turns out to be usable. A plant with current drawings and accessible panels moves quickly. One with drawings from three owners ago does not.
Is NFPA 70E enforced by OSHA?
No — and this gets stated incorrectly often enough to be worth being precise about. NFPA 70E is a consensus standard, not an OSHA regulation, so it is not enforced directly by federal law.
What OSHA enforces is 29 CFR 1910.132(d), requiring the employer to assess the workplace for hazards and determine whether PPE is necessary, and 29 CFR 1910.335(a)(1)(i), requiring that employees working where electrical hazards are present be provided with and use appropriate electrical protective equipment. OSHA accepts NFPA 70E as the nationally recognized consensus standard describing how that assessment is performed.
So the obligation is real and enforceable. NFPA 70E is the recognized method for meeting it, and the arc flash study is the documentation that you did.
The arc flash risk assessment must be reviewed at intervals not exceeding five years, and sooner whenever a change to the electrical distribution system could affect the results.
Who performs the study
Worth asking directly, because the answer varies more than people expect. Some firms are electrical contractors who broker the engineering to a subcontracted P.E. Some send technicians to collect data and have an engineer review it afterward.
At Phase Engineering the engineer who seals the report is the engineer who walked your plant. That is not a workflow preference. Incident energy results are only as good as the field data behind them, and the person accountable for the number should be the person who recorded the inputs.
Talk to the engineer who does the work
No sales team. You speak directly with a California licensed Professional Engineer about your equipment, your schedule, and what the study will actually cost.