Arc Flash Analysis Step-by-Step Instructions for Malaysia - Carelabs Malaysia

An arc flash study is a calculation, not an inspection. It answers one question at every point a worker may open a panel: how much thermal energy would reach them if an arc started while they stood there. The answer is expressed in calories per square centimetre at a stated working distance, and everything else — labels, PPE, work permits, approach boundaries — follows from that number.

Malaysia has no regulation that names arc flash directly. The duty comes from elsewhere. The Occupational Safety and Health Act 1994 places a general obligation on an employer to provide a safe system of work, and the Electricity Regulations 1994 govern how installations are operated and who is competent to work on them. DOSH assesses whether the hazard was identified and controlled. An employer who has never quantified incident energy has no defensible answer to that question, which is why the study matters well before an incident makes it urgent.

Step 1: Fix the scope before the data collection

Decide which equipment is in scope. In practice this is everything a person may work on or near while energised: LV main switchboards, MCCs, distribution boards, HV switchgear, and any panel where testing, racking or fault-finding happens live. Equipment that is never opened energised can be excluded, but that decision should be recorded rather than assumed.

Step 2: Build an accurate single line diagram

Most studies fail here rather than in the calculation. Drawings are reconciled against the installation as built, because a drawing set that predates the last two extensions will produce a confident, wrong answer. Transformer ratings and impedances, cable sizes and lengths, and switchgear nameplate data are all captured on site.

Step 3: Get the utility fault level from TNB

Incident energy scales with fault current and with arcing duration, so the source matters. A fault level letter from TNB for your point of connection gives both a maximum and a minimum value. Assuming a fault level from the transformer rating alone understates the duty on a strong supply and produces labels that are wrong in the unsafe direction.

Step 4: Run the short circuit study

Fault current at every bus is calculated to IEC 60909. This is a separate study, and it is a prerequisite rather than an optional companion — the arc flash calculation takes its bolted fault current directly from it. Motor contribution above roughly 50 kW is included, since large motors feed the first cycles of a fault.

Step 5: Establish clearing time from the protection

Arcing duration is the term with the most leverage over the final number. Halving the clearing time roughly halves the incident energy. Clearing time is read from the actual protective device characteristic at the calculated arcing current, using the settings the relays are really configured with — not the settings recorded on the schedule, which frequently differ after years of commissioning adjustments.

Step 6: Calculate incident energy to IEEE 1584

The IEEE 1584-2018 model is applied at each equipment location, using the electrode configuration, enclosure dimensions, gap and working distance that match the equipment being assessed. The standard's empirical model covers systems from 208 V to 15 kV and fault currents from 500 A to 106 kA, which spans the great majority of Malaysian industrial installations.

Step 7: Determine boundaries and PPE

The arc flash boundary is the distance at which incident energy falls to 1.2 cal/cm², the threshold for a second-degree burn on bare skin. PPE is then selected against the calculated energy at the working distance. Where energy exceeds the practical limit of arc-rated clothing, the correct response is to remove the exposure rather than to specify heavier PPE.

Step 8: Label, train and act on the findings

Labels are applied to each assessed panel showing incident energy, working distance, arc flash boundary and required PPE. The study is only useful once the people who open those panels understand what the label means, so training belongs inside the scope rather than after it. Any location that returns a high energy figure is a design finding, and the report should carry mitigation options for it.

What the study is not

It is not a one-off document. Fault levels change when TNB reinforces the network, when a transformer is uprated, or when embedded generation is added, and every one of those changes the number on the label. A study that describes a network you no longer operate is worse than no study, because it carries the authority of a calculation while stating the wrong energy.