What Causes In Rush Current and How to Limit It

In rush current (universally written as inrush current in datasheets) is the instantaneous, high-amplitude surge of electrical current that flows into a circuit the moment it is energized, primarily to charge empty capacitors or magnetize transformer cores. This transient spike fundamentally changes how you size upstream protective devices, forces the selection of higher-surge-rated rectifier diodes, and can cause severe voltage sags on shared DC buses that brownout nearby microcontrollers. Beginners commonly confuse it with short-circuit current (a catastrophic fault) or locked-rotor current (a mechanical motor stall), but inrush is a normal, expected physical behavior in a perfectly healthy circuit.

The Physics of the Startup Spike

When you first apply voltage to an uncharged bulk capacitor, its initial impedance is essentially zero. It acts like a dead short. The only things limiting the current are the Equivalent Series Resistance (ESR) of the capacitor, the resistance of the copper wiring, and the internal impedance of the power source.

Think of a bulk capacitor like an empty municipal water tower connected to a high-pressure main. When you first open the valve, water rushes in at maximum flow rate to fill the empty volume. As the tower fills and back-pressure builds, the flow rate naturally tapers off. In an electrical circuit, the 'flow' is current, the 'volume' is capacitance, and the 'back-pressure' is the rising voltage across the capacitor plates.

Because the initial resistance is incredibly low, Ohm's Law ($I = V / R$) dictates that the initial current will be massive. This spike typically lasts only a few milliseconds to a few tens of milliseconds, but that is more than enough time to melt silicon junctions, weld relay contacts, or trip magnetic breakers.

Worked Numeric Example: Sizing the Spike

Let's look at a typical 48V DC motor controller bus. You have a bank of electrolytic capacitors totaling 4700µF to smooth the PWM switching noise.

• Source Voltage (V): 48V DC
• Capacitance (C): 4700µF (0.0047 F)
• Capacitor ESR: 25mΩ (0.025Ω)
• Wiring & Source Impedance: 25mΩ (0.025Ω)
• Total Resistance (R): 50mΩ (0.050Ω)

At the exact microsecond the contactor closes, the capacitor voltage is 0V. The full 48V is dropped across the 50mΩ of parasitic resistance.

Your 48V system is drawing nearly 1000 Amps for the first ~235 microseconds (the time constant $ au = R imes C = 0.050 imes 0.0047$). If your upstream DC fuse is rated for 20A continuous but lacks the proper time-delay characteristics for this $I^2t$ energy let-through, it will blow instantly every time you turn the machine on.

Where You Meet This in Practice

You will encounter destructive in rush current in almost every power conversion discipline:

• Switch-Mode Power Supplies (SMPS): The bulk primary-side capacitor (often 100µF to 470µF at 400V DC) draws a massive spike from the mains rectifier the moment AC is applied.
• AC Induction Motors: 'Magnetizing inrush' occurs as the stator core establishes its magnetic field, often hitting 10x to 15x the full-load amperage (FLA) for the first half-cycle.
• LED Drivers: Constant-current drivers with large output filter caps will trip sensitive GFCI or AFCI breakers if multiple fixtures are switched on simultaneously.
• EV and Battery Systems: Connecting a high-voltage battery pack to an inverter without a precharge circuit will weld the main contactors shut due to the inrush into the inverter's DC link capacitors.

Real-World Scenario Walkthrough: The Blown Rectifier

The Setup: A hobbyist builds a 24V 10A linear power supply for a CNC router. The design uses a 20VAC toroidal transformer, a KBPC5010 bridge rectifier (rated for 50A continuous), and a massive 10,000µF filter capacitor.

1. The Numbers: The peak DC voltage after rectification is roughly $20VAC imes 1.414 = 28.2V$. The total circuit resistance (transformer winding resistance + wiring + cap ESR) measures about 100mΩ (0.1Ω).
2. The Event: The builder flips the mains toggle switch. There is a loud 'pop', a flash from the PCB, and the workshop 15A branch breaker trips.
3. The Outcome: Inspecting the board reveals the KBPC5010 bridge has cracked open. One of the internal diodes failed short, then overheated and shattered the epoxy case.
4. What Went Wrong: The builder sized the bridge for the continuous 10A load, ignoring the startup transient. The peak inrush was $28.2V / 0.1Ω = 282A$. While the KBPC5010 can handle 50A continuously, its non-repetitive peak surge current rating ($I_{FSM}$) for a single 10ms half-cycle is roughly 500A, and its $I^2t$ rating (thermal energy limit) was exceeded by the prolonged decay curve of the 10,000µF capacitor. The silicon junction melted before the capacitor reached full voltage.

Proven Methods to Limit In Rush Current

To prevent component destruction and nuisance breaker trips, you must intentionally add resistance during the startup phase. Here is how the three main techniques compare:

For most DIY and hobbyist builds under 500W, an NTC inrush current limiter is the undisputed choice. It is a negative temperature coefficient thermistor that starts with high resistance (e.g., 10Ω) at room temperature, limiting the spike. As current flows, it self-heats, dropping its resistance to a fraction of an ohm, allowing normal continuous operation with minimal voltage drop.

Frequently Asked Questions