LED Flicker: Where It Comes From and Why "Flicker-Free" Is Not a Marketing Word

Point a phone camera at a poorly driven LED and you will see dark bands rolling across the screen. That is flicker — a periodic modulation of light output — and it is the one defect that turns a well-specified lighting job into a pile of complaints. Knowing where it comes from is the first step to avoiding it.

Every LED driver converts AC mains to DC, and the quality of that conversion decides how smooth the light is. The simplest circuits rectify the mains and smooth it with a capacitor; if the capacitance is small, the output voltage — and therefore the current through the LEDs — sags between mains peaks, producing light that pulses at twice the line frequency, 100 Hz or 120 Hz depending on the region. Better designs use heavier filtering or active power-factor-correction stages that keep the current flowing continuously, which is why output ripple and flicker tend to be lower on the same products that carry high power factor. Dimming is the second source. If a driver dims by switching the LED current on and off at a few hundred hertz — low-frequency pulse-width modulation — the light follows the duty cycle, and the modulation becomes visible, especially in peripheral vision or during fast eye movement.

The industry measures this with two classic metrics. Percent flicker is (max − min) ÷ (max + min) × 100: a simple amplitude measure. Flicker index is more sensitive, weighting the shape of the waveform by dividing the area above the mean by the total area. On top of these, the stroboscopic effect — moving objects appearing to stutter — is assessed separately, in Europe commonly with the stroboscopic visibility measure, SVM, introduced in IEC TR 61547-1. Flicker and the stroboscopic effect together are called temporal light artifacts, and the term is useful because it keeps both failure modes in view at once.

Whether a given level of flicker is acceptable has been studied in detail. IEEE 1789-2015, the recommended practice for modulating current in high-brightness LEDs, is the reference most designers work to. Its low-risk guidance runs like this: between roughly 90 Hz and 1250 Hz, keep percent flicker below about 0.08 times the frequency — at 120 Hz that works out to about 10 percent; below 90 Hz the allowance shrinks further; and above about 1250 Hz, modulation depth is no longer considered a biological risk, which is why high-frequency PWM exists as a design strategy. The practical benefits of staying inside these levels are the ones people actually feel: fewer headaches in sensitive occupants, cleaner camera work in retail and broadcast environments, and no ghosting of rotating machinery in industrial spaces. For the small population with photosensitive epilepsy, flashing light in the roughly 3–70 Hz range is a genuine hazard, another reason aggressive low-frequency stroking of LEDs is bad practice in occupied spaces.

This is where a 0–10 V analog interface earns its keep. The dimming command is a steady DC level, not a train of pulses, so the driver is free to keep its power stage in continuous conduction at every level and reduce the LED current smoothly. That is the design intent behind flicker-free claims on architectural drivers: the dimming process itself introduces no low-frequency PWM artifacts, and the output filtering keeps mains ripple low. The honest caveat is that final flicker performance belongs to the whole chain — driver, module and how the two are matched — which is why responsible specifiers ask for measured values rather than trusting the word on the box.