Arc flash study services

Step 1 · the scope of work

What an arc flash study scope of work must contain

A loose scope is the single most expensive thing in this process. It does not save money at tender stage; it moves the money to a variation later, after you have already awarded and have no leverage. Nine things decide whether a bid is comparable.

1. The equipment threshold. IEEE 1584.1 guidance is that all equipment at 208 volts three phase and above is included. State it, and state any exclusion in writing rather than leaving it to interpretation.
2. The bus count, or how it will be established. If you know it, publish it. If you do not, say that the vendor must derive it from the single line diagram and declare it in the bid. Without this, no two prices mean the same thing.
3. The calculation standard. IEEE 1584-2018. Name it, because the retired 2002 method produces different answers and some bids still use it.
4. The data collection method. Who walks the plant, whether nameplates are photographed, whether trip unit settings are read at the relay, and who escorts. This is where scope quietly expands.
5. The single line diagram basis. State whether a current SLD exists and will be issued, or whether it must be reconstructed from the field. This is the largest cost variable on a brownfield site and it belongs in the tender, not the variation.
6. The utility fault data. Say who obtains the available fault current at the service. If nobody does, the study runs on a conservative assumption and every downstream result is inflated.
7. The label format and quantity. NFPA 70E format, the data fields required, material, and whether installation is in scope. Printing and installing are separable and should be priced separately.
8. The deliverable list. Incident energy report, reconciled SLD, short circuit and coordination results, label files, executive summary and action register. Anything unstated will be omitted.
9. The acceptance criteria. Who reviews, against what, and what constitutes completion. Include the requirement for a named signature and a conformance matrix at the front of the report.

The standards a scope should name

Name them explicitly rather than referring to applicable codes. Four documents carry the weight, and a bidder who cannot say which clause governs which deliverable has not read them.

Document	What it governs in your scope
NFPA 70E	Work practice. The risk assessment procedure, approach boundaries, PPE selection, energised work permits, label content and the review interval.
IEEE 1584-2018	The calculation. Arcing current, incident energy at the working distance, arc flash boundary. The 2002 method is retired.
IEEE 1584.1	The scope itself. Specification of scope and deliverable requirements, including the equipment threshold.
NFPA 70B	Equipment maintenance. Relevant because the condition of a protective device decides whether it clears in the time your calculation assumed. NFPA 70B compliance and arc flash results are connected, and a thorough auditor asks about both.

Together with OSHA 29 CFR 1910 these are the IEEE electrical safety standards and consensus documents a US arc flash tender is written against. Cite them by name in the scope and the bids you receive will be comparable.

Step 2 · cost

What drives the price, and why quotes differ so much

Two quotes for the same plant can differ by a wide margin simply because one counted 180 buses and the other counted 310, and neither said so. Ask every vendor for their bus count before you compare a single figure. If they cannot give you one, they have not scoped your plant.

What moves the count

• Assessed buses and panels. The primary driver, and the thing to agree first.
• Voltage levels and number of substations. More transformation stages mean more modelling, more coordination and more distinct results.
• Number of sites. Multi site programmes carry mobilisation per location.
• Scope beyond the calculation. Label printing and installation, floor boundary marking, qualified person training and the five year revalidation retainer sit outside the per label unit and should be priced separately so you can see them.

The four things that become a variation after award

• No current single line diagram. Rebuilding it from the field rather than verifying against it expands the walkdown considerably.
• Protection settings unavailable. Missing relay settings mean a return visit, or a conservative assumption that gives you a worse answer.
• No utility fault data. The source is assumed conservatively, inflating results at every downstream bus.
• Access and outage constraints. Equipment that opens only in a coordinated outage is sequenced against that window, not the study programme.

All four are visible during scoping if anyone asks. We ask, and we put the answers in the proposal so the basis of the number is on the page. Tell us your plant voltage and site count and a Chartered Engineer will come back with the scoped bus count.

Enquire Now

How to perform an arc flash study · the method your scope is buying

Six steps, in this order. A bid that cannot describe them is bidding on software output.

1. Field walkdown and data collection. Nameplates, CT and PT ratios, trip unit settings, cable runs, photographed and logged per equipment item.
2. Single line diagram reconciliation. Drift against the as installed plant logged and co-signed before any model is built.
3. System modelling and short circuit calculation. Bolted fault current at every bus, from verified source impedance and utility data.
4. Protective device coordination. Clearing time at each device · the largest single lever on the result.
5. Incident energy calculation to IEEE 1584-2018. Energy at each working position, arc flash boundary, approach boundaries and PPE category.
6. Labels, report and handover. Print ready NFPA 70E labels, the signed report with a conformance matrix, and a workforce briefing.

Step 3 · qualification

Who can perform an arc flash analysis

This matters because the calculation is the easy part. Any competent user can drive the software. What separates a defensible study from a decorative one is whether the inputs were verified, whether the assumptions were declared, and whether a named engineer is accountable for both.

The pattern to watch for is an assumptions register that disclaims the as-built drawings and then proceeds anyway. That produces a result which is precise, wrong, believed, and hanging on the wall.

Step 4 · comparison

How to choose an arc flash study company

Compare the bus count first, because that is the scope. Then apply four checks to a sample report from each bidder. You can do all four in under an hour, and they will separate the arc flash study companies bidding on engineering from those bidding on software output.

Check	What to look for
Who signed it	A named engineer on every section, not a stamp on the cover. Ask who drafted and who independently reviewed. One signature on a 200 page document is a formality, not accountability.
Were the inputs verified	The single line diagram co-signed against the as installed plant, nameplates photographed, CT and PT ratios confirmed, trip unit settings recorded.
Are the assumptions declared	Every conservatism stated so an auditor can trace each number to its source. Where utility data was unavailable, the estimate should be named as an estimate.
Is it audit ready	A conformance matrix at the front mapping each section to the standard it satisfies. An auditor should verify compliance in minutes rather than hunt through appendices.

What you should be receiving

Full anatomy of the document, section by section, is on the arc flash report page.

What the numbers on the label actually mean for scope

Two labels from one engagement, printed from one calculation. Reading them side by side is the fastest way to see why bus count alone does not tell you the whole story.

The difference is not the standard and not the software. It is the fault duty and the clearing time at that bus, which is why the coordination study belongs inside the same scope. Halve the clearing time and you roughly halve the incident energy, which is an engineering recommendation rather than a procurement one, and it is what a good report leads with. The arc flash boundary itself is the distance at which incident energy falls to 1.2 cal/cm², the onset of a second degree burn on bare skin. The limited approach boundary and the restricted approach boundary are different things again · shock boundaries set by system voltage rather than by incident energy · and all three are routinely confused on site. Your scope should require all three to appear on every label. The underlying method, the formula and a worked calculation example are covered on arc flash hazards.

DC systems are a separate scope line. Battery rooms, UPS strings and 800Vdc data centre plant are not covered by the AC equations and need a DC incident energy calculation, with UL listed switching equipment for the duty. Put it in the tender explicitly · see DC arc flash study.

Before you tender

If you are still deciding whether you need one

Because the duty is written as an outcome rather than a prescribed method, an employer who has performed no assessment has nothing to point to. The same architecture applies to fire risk assessment and machine guarding. And the interval is not open-ended: the arc flash risk assessment is reviewed at intervals not exceeding five years, and sooner on any major modification or renovation.

In practice the tender is triggered by a citation or inspection finding, an incident investigation, an insurer requirement at renewal, a corporate EHS standard, commissioning of new plant, or the five year review falling due. The recurring side of that · label currency, training cycles and the five year re-study · is handled as a programme rather than a repeat purchase. That is VB Arc360.

Where we work, and how long it takes

VB Engineering runs a United States entity with the team in Houston and Katy, Texas, supported by the analytical bench in Hyderabad. Field work, label installation and handover briefings run from the US side; modelling runs across time zones, which is why a study that would take eight weeks sequentially completes in six.

Six weeks end to end for a typical industrial site, eight for a multi substation facility. Walkdown two to three days on site, SLD reconciliation about a week, modelling and calculation two to three weeks, label production and installation two weeks. No plant shutdown is required for the analysis · only label installation needs brief access to each item, sequenced around your outage window.

Regions, sectors and site coverage are set out on the United States page.

The studies inside the same scope

Questions buyers ask before awarding