Inrush Current Limiting: NTC Thermistors and Active Bypass Circuits
The moment a driver is connected to the mains, the input capacitors are empty and behave like a short circuit. Current flows through the rectifier into the capacitor bank, limited only by the line impedance, the fuse and the equivalent series resistance of the components. The result is a current pulse that lasts a few milliseconds but can reach tens of amperes, far above the steady-state draw. This pulse is what trips breakers, welds relay contacts and stresses the input rectifier, and it is the reason practical drivers include some form of inrush limiting.
The classic solution is a negative temperature coefficient thermistor in series with the input. At room temperature the NTC has a relatively high resistance, perhaps five to fifteen ohms, which caps the initial current. As current flows, the thermistor self-heats, its resistance falls by an order of magnitude, and the steady-state loss becomes small, typically one to two watts. The device is cheap and simple, but it has a weakness: after a brief power interruption the thermistor is still hot and provides little limiting, so a rapid off-on cycle produces a much larger pulse than a cold start.
Active inrush limiting solves that weakness with a switch. A resistor or thermistor limits the initial current, and after a short delay a relay or a MOSFET bypasses it, removing the steady-state loss entirely. The delay is set by a timer or by a control IC that monitors the bus voltage and closes the bypass when the capacitors are charged. Some designs use a triac or a dedicated inrush controller with programmable timing. The advantage is repeatable limiting on every start, including hot restarts, at the cost of a few extra components.
Sizing involves the energy, not just the peak. The fuse must survive the inrush pulse without opening, which is why fuse selection considers the melting integral, the I squared t rating, as well as the steady current. The input rectifier and the bus capacitor must tolerate the pulse, and the branch breaker must not trip on its magnetic element. When several drivers share a circuit, the pulses add if they occur together, which is why large installations sometimes stagger the start-up of driver groups.
Measurement tells you what the datasheet means. A current probe with sufficient bandwidth, connected around the input conductor, captures the first ten milliseconds after switch-on. Testing at both 90 and 264 volts shows the range, because a higher line voltage produces a larger pulse for the same circuit. Comparing the measured peak with the datasheet value verifies the product, and comparing the pulse with the breaker trip curve explains nuisance trips that no one can reproduce afterwards.
For the PV100, the input is rated for 200-240 volts at 50 Hz, and inrush behavior is part of the product design. In the field, the practical rules are to avoid repeated rapid switching on the same circuit, to stagger start-up where many drivers share a breaker, and to check the breaker curve before blaming the driver for a trip. An inrush limiter that has failed typically shows as a blown fuse at every power-on, which is a diagnostic clue rather than a random fault.
A common field symptom is a breaker that trips only when a circuit is switched on after a short outage. The thermistor is still warm from the previous run, so its resistance is low and the pulse is large. Staggering the restart, or choosing a driver with active limiting, removes the symptom without changing the breaker.
The choice between NTC and active limiting is a cost and reliability decision rather than a performance one. A small driver may accept the thermistor and its two watts of loss, while a larger unit cannot afford the heat inside a sealed enclosure. Either way, the datasheet should state the inrush current, because the value is what the breaker and the wiring must tolerate.
Dielectric Strength and Hipot Testing: What the Isolation Rating Means
Thermal Runaway and Overtemperature Protection in LED Systems
Related Article