Nuisance Tripping and Driver Failures on a 347-V LED Circuit
A 60-pole, 347-V single-phase LED roadway lighting circuit uses 150-W luminaires with 0-10V dimming controlled by a central lighting control panel. Each 20-A breaker serves eight luminaires. Within 18 months, 11 drivers fail open-circuit, and the breaker trips intermittently at the dawn on command. At 20% dimming, three luminaires flicker visibly. The driver datasheet lists 50-A peak inrush for 1.5 ms, a 10% minimum dim level, and a 0-10V sink current of 2 mA. The control panel’s 0-10V output is rated 20 mA total and is wired with unshielded 18 AWG in the same conduit as the 347-V feeders. A site surge counter shows 14 events above 2 kV in the past year. The maintenance supervisor wants to replace the 20-A breakers with 30-A breakers and to remove the dimming feature entirely.
1. Evaluate the supervisor’s two proposed fixes and explain the likely root causes of the trips, driver failures, and flicker.
Read the worked answer
The 30-A breaker is not a sound fix. Steady-state current is only 8 × 150 W / 347 V ≈ 3.5 A, so the 20-A breaker is not tripping on overload; it is likely tripping on the magnetic instantaneous element during the dawn restart. Eight drivers at 50 A peak for 1.5 ms can present about 400 A, and a 20-A thermal-magnetic breaker may trip below that. A larger breaker may nuisance-trip less, but it may not be coordinated with the conductors or the driver withstand and does not address the cause. Removing dimming may reduce control-related flicker but will not stop surge damage or inrush trips. The 11 open-circuit driver failures fit surge events (14 above 2 kV) or overheating. The flicker at 20% is near the driver’s 10% minimum and is aggravated by unshielded 0-10V wiring in the same conduit as 347-V feeders. The 2 mA sink × 8 = 16 mA is within the panel’s 20 mA rating, but noise and voltage drop on the control pair remain suspect.
2. Specify a correction package and the commissioning measurements you would require.
Read the worked answer
Correct in layers. Stagger the dawn restart in groups of two or three drivers, and verify breaker coordination with the actual inrush waveform; use a breaker series or rating approved for high-inrush LED loads rather than simply upsizing. Add a listed surge protective device at the panel and at remote poles if the site exposure warrants it. Replace failed drivers with the specified model, and check driver case temperature and mounting. For the control circuit, use shielded twisted-pair 0-10V wiring separated from the 347-V conductors, bond the shield at one end, verify polarity, and confirm that each driver sees the commanded signal. Set the minimum dim level no lower than 10% and confirm the panel’s dimming curve. Commission with inrush capture, breaker trip test, control-voltage measurement at the farthest driver, flicker observation at 100, 50, 20, and 10%, insulation resistance, and surge-counter reset.
3. Is the 20-A breaker adequate for the steady-state load? Explain what that means for the troubleshooting direction.
Read the worked answer
The 20-A breaker is adequate for the continuous load. At 150 W per luminaire and 347 V, eight luminaires draw about 1,200 W, or 3.46 A; even with driver losses and the 125% continuous-load factor, the circuit is far below 20 A. The trips are therefore not an overload problem. Replacing with 30 A would mask the inrush issue and could reduce protection coordination; it is not the right first response. The better solution is to control inrush and verify the breaker’s magnetic trip characteristics.