Reducing Fault Current
Source impedance, current-limiting devices, and topology choices.
Three levers reduce arc-flash energy. This lesson is about the first — lowering the available bolted fault current at the bus where the work happens. Lesson 7 walks through the second (reducing clearing time); the third (increasing working distance via barriers, remote operation, or system reconfiguration) is mostly a field-operations question and doesn’t get its own dedicated lesson here.
Lowering Isc isn’t as immediately effective as lowering clearing time — arc-flash energy scales roughly linearly with clearing time but only sub-linearly with Ibf. But it’s the right lever to reach for when the equipment is over-dutied (Isc exceeds breaker AIC), or when the upstream protective device is fixed (it’s a fuse and you can’t change its TCC).
Lever 1: source impedance
The simplest approach — pick a transformer with a higher %Z.
The starting state — 2000 kVA / 5.75 % Z — lands around 41 kA at the bus. That’s right at the limit of what 42 kA-rated 480 V switchgear can interrupt. A small change in utility availability or motor contribution might push it over. The arc-flash label panel below reads around 9 cal/cm² (Cat 3) at the default 200 ms clearing time.
Try this: raise the transformer %Z from 5.75 % to 7.5 %. Isc drops to about 31 kA — comfortably under 42 kA AIC, and the arc-flash label tracks down with it to about 7 cal/cm², dropping from Cat 3 into Cat 2 — about 22 % less energy at the worker, a meaningful PPE margin when the clearing time can’t be improved further.
The tradeoff: 7.5 % Z means more voltage regulation under normal load. A motor starting at full kVA inrush will see a bigger voltage dip. In some installations that’s a deal-breaker (precision manufacturing, sensitive electronics); in others it’s irrelevant.
Lever 2: current-limiting fuses
A class-L or class-J current-limiting fuse in the upstream feeder caps the let-through current at high fault levels. The fuse goes into current-limiting at a few-tens-of-times its rating; once there, its time-to-clear is sub-cycle (half a cycle or less), which truncates the peak fault current waveform before it reaches the full prospective Isc.
For the arc-flash calculation specifically, the clearing time is the protective device’s actual time to open — and current-limiting fuses are exceptionally fast in their let-through region. Many arc-flash studies show incident energy under 1.2 cal/cm² on bus segments protected by an upstream current-limiting fuse, even at high available Isc, simply because the fuse clears in < 8 ms.
The catch: current-limiting fuses don’t help if the fault current is below their threshold. An 800 A class-L fuse goes current-limiting at roughly 20× rating ≈ 16 kA. Below that, it behaves like an ordinary fuse with second-scale time-current characteristics. So a moderately-sized fault in the 1–15 kA range could still produce high arc-flash energy.
Lever 3: cable / reactor impedance
The bus you’re worried about is downstream of something — a feeder cable, a tie reactor, sometimes a long isolated-phase bus run. Whatever’s between the source and the work point adds series impedance and drops Isc at the work point.
Try this in the widget: turn the cable run on, set length to 100 ft of 750 kcmil copper, 4 parallel sets — realistic sizing for a 2000 kVA / 480 V feeder. Bus Isc drops by ~3 kA. The further away the work point is from the bulk source, the smaller the available fault.
This is why arc-flash labels at downstream panelboards often show lower energy than at the main switchgear they’re fed from. A worker on a 200-ft-distant panel is working against a smaller fault and gets a lower-category label.
Cable Z is mostly reactance at the gauges and lengths typical of 480 V feeders, so it adds proportionally to the X side of the bus X/R ratio. The change in X/R can have second-order effects on asymmetrical peak duty.
Series reactors are the deliberate version of this — installed specifically to limit Isc on a bus that would otherwise be over-dutied. Common in industrial substations where adding a higher-%Z transformer isn’t an option (motor starting voltage drop) but a small reactor on the feeder is.
Lever 4: topology
Sometimes the right answer isn’t changing any impedance — it’s changing what’s connected to what.
- Open the tie breaker on a double-ended substation during work on one of the two bus sections. The faulted bus is fed from only one transformer instead of two — Isc roughly halves.
- Remove the motor contribution. If a large motor on the bus is the dominant motor contribution, securing it for the duration of the work cuts a chunk of Isc.
- Reconfigure to a smaller transformer. Some industrial loads can be temporarily fed from a smaller backup transformer with higher %Z and lower Isc for the duration of the maintenance.
These are operations decisions, not design decisions — usually written into the energized-work procedure for the equipment.
When this lever isn’t enough
Reducing Isc has diminishing returns. Once you’re under the breaker AIC, the arc-flash energy is still mostly set by clearing time, not fault current. A 25 kA bolted fault at 200 ms (~5.8 cal/cm² / Cat 2) vs. a 15 kA fault at the same 200 ms (~3.7 cal/cm² / Cat 1) — that’s a 40 % reduction in Isc for a single category drop in PPE.
The next lesson goes after the bigger lever: clearing time. Cutting the same 200 ms down to 50 ms drops the same 25 kA fault from Cat 2 (~5.8 cal/cm²) down to Cat 1 (~1.5 cal/cm²) — without changing a thing about the system impedance.
What’s next
Lesson 7 picks up the TCC tutorial directly. The same upstream LVPCB, same downstream branch, same coordination story — but now we’re trading off coordination and arc-flash energy. ARMS, RELT, ZSI, and NEC 240.87 enter the picture.