Calculating Isc at a Bus
Per-unit method walked through with an interactive bus calculator.
The simplest way to compute fault current on a bus is the per-unit method: pick a base MVA, convert every impedance in the path to per-unit on that base, add them up (series for the source path, parallel for motor contributions), and the bus Isc is the base current divided by the total per-unit impedance.
The math is voltage-agnostic — that’s the whole point. A 5.75 % Z transformer is 5.75 % Z whether you measure it from the primary or secondary side, so per-unit lets you walk through a one-line that changes voltage levels without rewriting Z every time.
The one-line we’ll model
The widget below shows the canonical industrial supply path: utility behind a transformer, transformer feeding a 480 V switchgear bus, panelboards downstream. Click the bus to read its short-circuit and arc-flash numbers; switch to the double-ended topology to see how a tie-breaker between two paralleled transformers roughly doubles Isc at either main when closed.
We’ll also let you toggle in a motor contribution at the bus — a lumped block of running motors that pump current into a downstream fault for the first few cycles — via the SC calculator’s sliders.
The interactive one-line
Start position is a 1500 kVA / 5.75 % / 480 V transformer fed from a 500 MVA utility primary. You should read about 30 kA at the bus.
Try this — change one thing at a time
Bigger transformer
Drag the transformer rating up to 2500 kVA. Watch Isc climb to
~50 kA. Larger transformer = smaller per-unit impedance on the system
base = more fault current. The infinite-bus shortcut version of this
is Isc ≈ kVA / (√3 · kV · %Z/100) — direct proportionality with kVA.
Higher impedance transformer
Hold the kVA where it is and increase %Z from 5.75 to 7.0. Isc drops to ~25 kA. This is why power engineers sometimes specify higher impedance transformers — to deliberately limit available fault current on a bus that would otherwise be over-dutied for its breakers. The tradeoff is voltage regulation: higher %Z means more voltage drop under load.
Long cable run
Turn the cable toggle on. Set length to 100 ft, gauge 500 kcmil, copper, 4 parallel sets — that’s a realistic feeder for a 1500 kVA / 480 V bus carrying ~1800 A FLA. Isc drops by ~3 kA. The cable Z adds in series with the source path. Smaller gauge, longer run, or fewer parallels → more added Z → less Isc at the far end. Try 2 AWG copper with 1 set to see how aggressively a small cable knocks Isc down — but note that 2 AWG can only carry ~115 A, so it’s not a realistic feeder for this bus.
A practical consequence: the fault current at a panelboard 200 ft down a small feeder is meaningfully lower than at the main switchgear. Many engineers compute the Isc at the panel, not the main, when selecting branch-breaker AIC ratings.
Motor contribution
Turn motor on. Set HP to 800. Isc climbs by 3–4 kA. The motors behave like generators behind their own reactance during the first few cycles and feed the fault. On a heavily motored industrial bus, the contribution can be 20–40 % of the transformer’s, which sometimes pushes equipment over its AIC rating without anyone realizing.
The motor contribution decays over the first ~5 cycles as rotors slow down — production tools model this with separate “first-cycle” and “interrupting” Isc values. For now we use the initial symmetric contribution everywhere, which is conservative.
A real coordination study uses this number
The Isc you just computed is the x-axis maximum on the TCC plot back in Module 1 of the coordination tutorial. It’s also the Ibf input to the arc-flash incident-energy model in the next module. Every downstream decision — breaker AIC, cable size, fuse choice, arc-flash PPE — references this number.
Caveats worth knowing
- The per-unit method assumes balanced 3-phase steady-state. For unbalanced faults (L-L, L-G), you need symmetrical components — positive, negative, and zero sequence networks. Production tools do this automatically; manual hand-cranks for L-G are tedious.
- Motor contribution beyond the first ~5 cycles requires per-motor modeling with subtransient / transient / synchronous reactances. Our lumped 4× FLA is fine for first-cycle / momentary duty but overstates real behavior at longer time scales (motor contribution decays as the rotors slow, so a constant 4× over-counts it later in the event).
- Cable Z here uses NEC Chapter 9 Table 9 approximations. Real designs use manufacturer data when the cable choice is critical (parallel runs, isolated phase bus, etc.).
What’s next
Module 2 puts Isc to work. The same bus, the same fault, but now the question is: if an arc starts at that bus and the upstream device takes 100 ms to clear, how much energy reaches a worker standing 18 inches away?