Designing an Active Inrush Current Protection Circuit for DC Loads

The Core Problem: Why Power Supplies Trip and Fuses Blow

An inrush current protection circuit limits the initial surge of current when charging large capacitive loads, preventing breaker trips, blown fuses, and severe voltage sags. When you apply 24V to a raw 4,700µF capacitor bank, the initial current is limited only by the parasitic resistance of your wires and the ESR of the capacitors. Using the formula I = C × (dV/dt), if the voltage ramps up in just 1 millisecond, the inrush spike hits 112 Amps. This will instantly trip a 10A breaker, weld relay contacts, or trigger the over-current protection (OCP) on your bench supply.

The direct answer for high-power or frequently-cycled DC loads is an active P-channel MOSFET soft-start topology. Unlike passive thermistors, it drops almost zero voltage at steady state and resets instantly.

Topology Showdown: NTC Thermistor vs. Active MOSFET

Before wiring components, we must choose the right approach. The two most common topologies are the passive Negative Temperature Coefficient (NTC) thermistor and the active MOSFET RC soft-start.

Why this topology over the alternative? If you use an NTC in a system that is turned off and back on quickly (like a motor controller resetting from a fault), the thermistor is still hot. Its resistance remains low, and it provides zero inrush protection on the second cycle. The active MOSFET circuit relies on capacitor charge times, which are independent of ambient temperature and reset immediately upon power removal. For any DC application where reliability and rapid cycling matter, the active topology wins.

Active P-Channel MOSFET Soft-Start: Topology and Node Map

The active inrush current protection circuit relies on an RC network to slowly ramp the gate voltage of a P-channel MOSFET, forcing it to operate in its linear (ohmic) region momentarily before fully saturating.

The Circuit Path:
The Source pin of the P-FET connects to VIN. The Drain pin connects to VOUT. A pull-down resistor (R1) connects from the GATE to GND. A timing capacitor (C1) connects directly between the GATE and the SOURCE (VIN). Finally, a Zener diode (D1) is placed between GATE and SOURCE to protect the gate oxide.

When VIN is applied, C1 initially acts as a short, pulling the GATE up to VIN. The Gate-to-Source voltage (Vgs) is 0V, and the MOSFET is completely OFF. As C1 slowly charges through R1 to GND, the GATE voltage drops. Vgs becomes increasingly negative, slowly turning the MOSFET on and ramping VOUT up to VIN.

Design Walkthrough: Sizing the IRF4905 and RC Network

Let's design a concrete circuit for a 24V DC system driving a load with 4,700µF of input capacitance and a steady-state draw of 5A.

1. Selecting the MOSFET (Q1)

We need a P-channel MOSFET with a Vds rating well above 24V and a continuous drain current (Id) above 5A. The IRF4905 (TO-220 package) is the workhorse here. It features a Vds of -55V, an Id of -74A, and an Rds(on) of just 20mΩ. At 5A, steady-state power dissipation is a negligible I²R = 25 × 0.020 = 0.5W.

2. Sizing the Gate Protection (D1)

The IRF4905 has a maximum Vgs rating of ±20V. If we apply 24V directly across the gate and source, we will punch through the gate oxide and destroy the part. We place a 15V Zener diode (e.g., BZX55C15) between GATE and SOURCE (cathode to SOURCE, anode to GATE). This clamps the maximum Vgs to a safe -15V.

3. Calculating the RC Time Constant (R1 and C1)

We want the soft-start ramp to take roughly 300ms to keep inrush under 1A. The turn-on time is governed by the time constant τ = R1 × C1. The MOSFET begins conducting heavily around 3τ.

• Choose C1 = 1µF (ceramic or film, rated for 50V).
• To get a 300ms turn-on (where 3τ ≈ 300ms, so τ = 100ms):
  R1 = τ / C1 = 0.1s / 0.000001F = 100,000Ω.
• Select R1 = 100kΩ (1/4W standard resistor).

With these values, the inrush current is limited to roughly I = C × (dV/dt) = 4700µF × (24V / 0.3s) = 0.376A. We have successfully reduced a 112A spike to a safe 376mA ramp.

Behavior Matrix: What Happens When Components Shift?

Understanding how component tolerances and variations affect the circuit is critical for production or field repairs.

Failure Mode Analysis: Shorts, Opens, and Extremes

A robust design must fail safely. Here is what breaks at the extremes and how the circuit responds.

• R1 Opens: The GATE has no path to GND. C1 charges instantly to VIN. Vgs remains 0V. The MOSFET never turns on. Result: VOUT stays at 0V. Safe failure.
• C1 Shorts: The GATE is instantly pulled to GND via R1. Vgs immediately hits -24V (clamped to -15V by D1). The MOSFET turns on instantly. Result: Inrush protection is defeated, full 112A spike hits the load. The fuse may blow, but the load survives.
• Q1 Shorts (Drain-to-Source): A common MOSFET failure mode is a shorted channel. Result: VIN passes directly to VOUT. The load continues to operate normally, but inrush protection is permanently bypassed until Q1 is replaced.
• D1 (Zener) Shorts: The GATE is clamped directly to the SOURCE. Vgs is forced to 0V. Result: MOSFET never turns on. VOUT = 0V. Safe failure.

Breadboard Testing Protocol: Step-by-Step Verification

Do not just wire this up and flip the switch. Follow this bench protocol to verify the RC curve without risking your load or power supply.

1. Prep the Bench Supply: Set your DC power supply to 24.0V. Engage the current limit (OCP) and set it to 1.5A. This ensures that if your circuit fails, the supply will fold back rather than deliver a destructive spike.
2. Wire the RC Network First: With power OFF, insert the IRF4905, R1 (100kΩ), C1 (1µF), and D1 (15V Zener) into the breadboard. Double-check the Zener orientation: the stripe (cathode) must face the SOURCE (VIN) pin.
3. Connect a Dummy Load: Do not connect your expensive microcontroller or motor driver yet. Connect a 10Ω, 10W power resistor between VOUT and GND to simulate a steady load, alongside a 100µF test capacitor.
4. Probe the Nodes: Connect Channel 1 of your oscilloscope to VIN and Channel 2 to VOUT. Set the timebase to 100ms/division.
5. Apply Power and Observe: Turn on the supply. You should see VIN step up instantly to 24V. VOUT should remain at 0V for roughly 50ms, then ramp up smoothly, crossing the 12V mark around 150ms, and settling at 24V near 350ms.
6. Verify Steady-State Drop: Switch the oscilloscope to DC coupling and measure the voltage difference between VIN and VOUT. With a 10Ω load drawing ~2.4A, the drop across the IRF4905 should be roughly 2.4A × 0.020Ω = 48mV. If you see a drop greater than 200mV, your MOSFET is not fully saturating—check your R1 and GND connections.

The Final Verdict: Decision Tree for Inrush Protection

Use this decision matrix to finalize your component selection for your next PCB or panel build.

For the vast majority of DIY robotics, amateur radio, and custom 12V/24V solar battery banks, the IRF4905 active inrush current protection circuit provides the best balance of cost, reliability, and zero steady-state power loss. By explicitly controlling the dV/dt ramp with a 100kΩ/1µF network, you eliminate the guesswork and protect both your power supply and your downstream electronics from catastrophic surge events.

References and further reading:
All About Circuits: Inrush Current and How to Limit It
Texas Instruments: Hot-Swap and Inrush Current Limiting Overview