Every method that reduces starting current does it by reducing the voltage the motor sees. Current falls in proportion to voltage; torque falls with the square of it. That single asymmetry is why soft starting is not free, and why the answer to “the breaker keeps tripping” is sometimes “then it will not start your load either”.
The breaker is only one constraint. Drawing several times full-load current for a few seconds pulls the supply voltage down for everyone on it — and because torque follows voltage squared, a deep dip can leave the motor unable to finish the very start that caused it.
At the instant of switch-on the rotor is stationary, so the rotating field sweeps past it at full speed. Electrically the machine looks like a transformer with a shorted secondary: low impedance, enormous current, and a power factor down around 0.2. As the rotor accelerates the slip falls, the impedance rises, and the current collapses to its running value.
For a three-phase motor, I = P / (√3 · V · PF · η). The efficiency term is there
because the plate rating is shaft output; the motor draws more than that. A 7.5 kW
400 V motor at 0.85 power factor and 90% efficiency draws 14.2 A running, and
92 A starting at 6.5 times.
Reduce the terminal voltage to a fraction a of nominal and the motor draws a times the current but produces a² times the torque. Star-delta is the special case everyone remembers: reconnecting the windings gives exactly one third of both. That symmetry is a property of the delta reconnection, not a general rule — a solid-state soft starter set to 300% draws more current than star-delta and delivers less torque.
A thermal-magnetic breaker has an instantaneous magnetic trip set at a multiple of its rating: 3–5× for type B, 5–10× for C, 10–20× for D. A motor circuit needs the magnetic trip above the inrush and the overload protection set to the running current — which is why motor protection is normally two devices, or one device designed for the job.
Because the two are protecting against different things. The thermal element responds to sustained overcurrent and is happy with a two-second inrush. The magnetic element responds instantly to a multiple of the rating, and a 6–8× inrush lands squarely in the type B window and often in the type C one. The fix is a device whose magnetic trip clears the inrush — a type C or D, or a motor circuit breaker — with separate overload protection sized to the running current, not a bigger breaker.
Rarely, on new work. It is cheap and it has no electronics, but it gives you a third of the torque, an open transition that produces a fresh current spike partway through, and no control over the ramp. A soft starter costs more and does the job properly; a VFD costs more again and pays it back in running energy if the load is a pump or fan. Star-delta survives mostly on existing installations and on unloaded starts — a compressor that unloads for starting, for instance.
As long as it takes the load to reach speed, which is a mechanical question, not an electrical one. An unloaded motor is up in well under a second; a large fan with real inertia can take fifteen or twenty. That duration is what decides whether the motor overheats, because a stationary rotor has no cooling and is absorbing several times its rated power. Repeated failed starts are far more damaging than one long one.
Yes, and more thoroughly than anything else, because it lowers the frequency along with the voltage. Keeping the volts-per-hertz ratio constant keeps the magnetic flux at its design value and the slip small, so the motor produces full torque or better while drawing little above its running current — typically 1.1 to 1.5 times. The catch is that the drive itself has an input inrush, it needs harmonic and EMC consideration, and it is the most expensive answer.
A standard cage induction motor with a locked-rotor current you supply, a stiff supply apart from the transformer impedance you enter, and reduced-voltage methods behaving ideally. Not modelled: the speed–torque curve of the motor or the load, so the page cannot tell you whether a given method will actually accelerate a given machine; rotor heating and permitted starts per hour; cable impedance and its own volt drop; the DC transient in the first cycles, which can briefly double the peak; open-transition spikes in star-delta changeover; drive input inrush and harmonics; or soft-starter ramp shaping.
No changes to this tool’s own behaviour since the earliest archived release (v1.22). The full history for the site is in the changelog.