Transformer energising inrush current (magnetising current) — why the primary fuse can't simply be sized on rated current

The guide on the main utility fuse versus the first installation breaker covers selectivity between successive protective devices in general. This article covers a specific case where a normal, fault-free action — switching in a transformer — can already draw enough current to trip an incorrectly sized primary fuse or breaker.

The phenomenon: magnetising current at switch-on

When an unloaded transformer is switched in, its magnetic core must be magnetised from zero. Depending on the exact instant of switching on the voltage waveform, and on the remanent flux left in the core from the previous switch-off, the core can temporarily enter saturation. In that saturated region the relationship between current and magnetic flux is strongly non-linear, so the primary winding draws a much larger current than the normal, continuous magnetising current — this is the inrush current.

Magnitude and decay of the peak

The inrush current of a power transformer can reach roughly 8 to 12 times the rated primary current at switch-on, concentrated in the first half-cycle (on the order of 0.01 second at 50 Hz). After this first, highest peak, the amplitude gradually decreases over the following cycles, until the transformer returns to its normal, much smaller continuous magnetising current after roughly 1 second. This peak is not a fault — the transformer and the connected installation are operating normally — but it is still a brief, significant current that must not cause the primary fuse or circuit breaker to trip unnecessarily.

Consequence for protective device selection

A primary fuse or breaker chosen purely on the basis of the rated transformer current and the required short-circuit protection can trip unnecessarily on every switch-on of the transformer if its time-current characteristic doesn't give enough margin relative to the inrush curve. To avoid this, the time-current characteristic of the chosen device must sit above the transformer's inrush curve: sufficiently high at the very short time of the first peak, and still sufficiently high for the somewhat slower-decaying current during the following second.

Note: the exact inrush curve is transformer-specific (rating, core design, switching instant, remanent flux), and the resulting protective device selection is a time-current coordination exercise between the transformer manufacturer's data and the time-current characteristic of the chosen fuse or breaker — this article covers the principle, not a ready-made calculation formula for every specific type.

Why this isn't a single, predictable value

Because the peak height depends partly on the incidental switching instant on the voltage waveform and on the remanent flux left in the core from the previous switch-off, the inrush peak is not identical on every switch-on: one time the peak may sit close to the upper end of the stated 8-12× order of magnitude, another time considerably lower. A protective device sized only just above an average inrush value can therefore still trip unnecessarily on an unfavourably timed switch-on.

Practical relevance

When selecting a primary fuse or installation breaker for a transformer, its time-current characteristic must be explicitly plotted against the inrush curve provided by the manufacturer (or otherwise determined) — not just against the rated operating current and the required short-circuit protection. A fuse or breaker that skips this step can cause an otherwise healthy transformer to trip on nearly every switch-on, which is particularly disruptive for transformers that are switched in and out regularly.

Common mistakes

Sizing the primary fuse or breaker purely on rated operating current and the short-circuit protection requirement, without checking its time-current characteristic against the transformer's inrush curve.
Treating a single inrush value as an absolute maximum without accounting for the spread caused by switching instant and remanent flux — a device sized just above an average value can still trip on an unfavourable switch-on.
Mistaking the inrush peak for an actual fault when analysing an unexplained trip right after switching in a transformer, instead of first checking the time-current coordination.
Not accounting for repeated switching (for example a transformer that is periodically taken out of service) — every individual switch-on brings its own inrush peak, with its own, variable remanent-flux starting condition.