Selection & sizing

How many drivers fit on one circuit

Steady-state current is a quarter of the breaker rating and the circuit still trips on switch-on. This guide gives the comparison rule between inrush peak and the breaker magnetic trip threshold, the B/C/D curve conversion, a worked 16 A type C example, plus on-site measurement and an acceptance checklist.

Last updated：2026-10-31

Steady-state current sits at a quarter of the breaker rating and the circuit still trips the moment you switch it on. That is not an undersized breaker and it is not a faulty driver. What trips a breaker is the inrush of the first few milliseconds, not the running current. Once you put a number on it, circuit counts stop being guesswork.

The one-sentence rule: compare peak against threshold, not watts against rating

N_max = (k x I_n) / I_peak

k       lower bound of the breaker magnetic trip band (B 3, C 5, D 10)
I_n     breaker rated current (A)
I_peak  cold-start inrush peak of one driver (A)

Use the lower bound. The upper bound is where it might not trip; the lower bound is where it will not trip.

Three quantities define inrush

1. Peak I_peak: after the rectifier the bulk capacitor looks close to a short. Peak depends on loop impedance, and it is highest when the switch closes at the voltage peak - closing at a zero crossing can cut it by a large factor
2. Duration: typically 0.2-5 ms before it decays to the running value. The shorter it is, the less likely the breaker is to act
3. How many units start together, N: the only variable you control directly

Topology moves the peak a long way: units with an NTC limiter or soft start can sit at 5-15 times running current, while unprotected designs reach tens of times. Always read the test conditions next to the published inrush figure - voltage, phase angle, cold or hot state.

Running current is what governs long-term heating, not tripping:

I = P / (U x PF)
e.g. 100 W, 230 V, PF 0.95  ->  I about 0.46 A

Worked example: 100 W drivers on a 16 A type C circuit

That is the counter-intuitive part: 11% of rating and it still trips. Published guidance for comparable mid- to high-power drivers quotes 2 units on a 16 A/230 V type B circuit and 3 units on type C (check the current manufacturer datasheet), which lands in the same range as the formula - a useful cross-check.

Breaker curve conversion

Magnetic trip bands are conventionally B 3-5x In, C 5-10x In and D 10-20x In under IEC 60898-1 (check the actual product data and local rules). The lower bound is the k in the formula:

Why the formula is only an upper bound: three real-world limits

1. Peaks do not add arithmetically: cable impedance, phase angle and the exact closing instant all reduce the measured total below N x I_peak. So the formula is conservative
2. A breaker responds to energy, not only peak: a very short pulse may pass the first time, but the thermal-magnetic mechanism has already moved, and the second or third attempt trips more easily. That is the mechanism behind "it was fine the first time"
3. Cold is worst: re-closing straight after a trip leaves the input capacitor partly charged and the peak is smaller. A fully cooled unit gives the highest peak. Test from cold; a hot retest flatters the design

Measurement: from the simplest to the quantitative

Simplest: switch the full circuit on ten times from cold and note whether it trips, and on which attempt. A single trip means the margin is gone.

Quantitative: put a low-value shunt resistor in series (size the resistance and power rating for the circuit current and keep a safe working distance) and convert the scope voltage drop to current, or use a clamp meter with peak-hold (inrush) and convert per the instrument manual. Record four things: single-unit peak, N-unit peak, whether it tripped, and the breaker model and curve.

What to do after it trips, cheapest first

1. Stagger the start: split the circuit into groups with 10-50 ms between them. Cheapest option by far - cascaded contactors or relay groups
2. Split into more circuits: extra ways are rarely the dominant cost, so do this early
3. Choose units with an NTC or soft start: ask the manufacturer for a measured inrush curve, or at least a figure with test conditions. A bare multiple is not enough
4. Raise the breaker curve: B to C, with the cable and fault-level checks above
5. Fit a lighting contactor: contacts rated for the inrush of LED drivers, not for the running current - a circuit that runs at 1.83 A still needs contacts that survive a hundred-amp make

Acceptance checklist

1. Take inrush peak, test conditions and duration from the datasheet; if duration is missing, assume 1 ms
2. Work out running current per unit: I = P / (U x PF)
3. Confirm breaker model, rating and curve; if the curve is unknown, design for type B
4. Apply N_max = (k x I_n) / I_peak with the lower bound of k and round down
5. Switch the full circuit on ten times from cold and require zero trips
6. File the measured peak together with the breaker data for the next time the circuit is extended

Deciding circuit counts at specification stage is far cheaper than fixing it on site. High-power units in the H series such as H01 and H19, and K01 in the K series, all go through the same formula - do not estimate circuit counts from running power, because that number is optimistic in a dangerous way. Check power and current conversions with the LED driver sizing calculator; for DC-side cable size and voltage drop see the 12 V/24 V voltage drop and cable size table, and for other causes of flicker after switch-on see the LED flicker troubleshooting guide.

© 2026 YFX. Figures are for selection purposes. Always follow the datasheet and local wiring regulations for final installation.