Arc Flash Study Cost: $3,500–$18,000 per Facility (2026 Prices)

A full arc flash engineering study costs $3,500 to $18,000 per facility, per pricing published by Zech Engineering (April 10, 2026). That range is wide because the study is mostly labor, and the labor is mostly set by your facility, not by the engineering firm's rate card.

This page breaks down where the money goes, what the per-panel math looks like, when paying for a study is the right call, and when it is more than you need. Prices below are dated and linked to their sources at the bottom.

Last reviewed: July 18, 2026.

The per-panel math

Studies are quoted per facility, but what you are really buying is a per-panel result. Dividing the sourced facility range by typical equipment counts — this is plain arithmetic on the $3,500–$18,000 range, not a separate price survey:

Two honest observations from that table. First, the per-panel price drops fast with scale, which is why firms like big facilities. Second, the small-shop end is where a study hurts most: $3,500 to label ten panels is a real number for a ten-person contractor.

What drives the price

Five variables decide where you land in the range:

1. Equipment count and facility size. Every bus, panel, MCC, and disconnect needs data collected, modeled, and labeled. This is the dominant variable.
2. Data collection effort. Jim Phillips, writing in EC Magazine, calls data collection potentially "a significant portion of the overall study effort" and warns against "data paralysis." If your one-line diagrams are current and your nameplates are legible, the study is cheaper. If the engineer has to reconstruct the system from panels with missing labels, you pay for detective work.
3. Utility fault-current data. The study needs available fault current from your utility, and getting it is the notorious bottleneck. In a Mike Holt's Forum thread on study workflow, engineers report months-long waits — and report that for services at 480 V and below, meaningful utility fault-current data arrives less than 10% of the time. One engineer described filing a state utility commission complaint to force the data out. Delays here stall the whole project and can change the fee.
4. The wrong first answer problem. The first number a utility sends is often the infinite-bus value from the transformer nameplate, which overstates the available fault current at your service. A NEMA technical paper warns that incident-energy calculations run on conservatively high fault-current values can underestimate the real arc-flash hazard. A competent firm chases the real number; that chase is billable.
5. Coordination and documentation scope. If the study includes protective-device coordination, updated one-lines, and a formal report for your insurer, it costs more than a label-only scope.

Software prices, for context

Some buyers compare the study quote against doing the analysis in-house with engineering software. Verified July 2026 prices:

These are tools, not studies. They produce numbers; they do not collect your field data, stamp anything, or stand behind the result. The $995-and-up tiers earn their price largely on one feature: automatic protective-device clearing times from a device library — the input most likely to be wrong when it is guessed.

When a study IS the right spend

Be suspicious of any page — including ours — that talks you out of a study you actually need. Pay for the engineering firm when:

• An AHJ, insurer, or customer requires a PE-stamped report. No software output substitutes for a stamp.
• Your system is outside calculation-tool scope. Systems above 15 kV, DC systems, and complex multi-source or multi-mode arrangements need full system modeling, not a per-panel calculation.
• You need a coordination study. Selective coordination and TCC work are engineering products.
• Labels show 40 cal/cm² or more. At that energy level the answer is usually engineering — current-limiting devices, maintenance settings, relay work — not a sticker.
• Your documentation is gone. If nobody can say what is upstream of a panel, a calculator cannot help you yet; the field data collection is the study.

When a per-panel calculation covers you

For a small contractor or a small facility labeling 5 to 50 panels, the full study is often more than the code asks for. NFPA 70E accepts an incident-energy analysis, and IEEE 1584-2018 defines that calculation for 208 V to 15 kV equipment. If your panels sit inside that range and you can supply the two hard inputs — available fault current and protective-device clearing time — a per-panel calculation gives you the label values without a facility-wide engagement.

That is exactly what our free arc flash calculator does: IEEE 1584-2018 math with every intermediate value shown, formatted onto an NFPA 70E §130.5(H) label. What it does not do is just as explicit: no engineering stamp, no field data collection, no coordination study. If you are missing the fault-current input, read what is available fault current. For what the resulting label must contain, see arc flash label requirements.

Interactive cost estimator — planned

We are building an interactive estimator: panel count and facility profile in, an honest cost range out, with the study-vs-calculator decision spelled out per case. It is not live yet. Until it ships, the sourced ranges and the per-panel table above are the honest version — no fake precision, no "instant quote" form that is really a lead magnet.