PNNL

An office worker sitting in a cubicle under an LED light mounted in the ceiling shouts to no one in particular, “That flicker is giving me a headache!” The office worker one cube over says, “You don’t know what you’re talking about! The light is fine!”

There, in a nutshell, is the problem with certain LEDs and their electronics: they can produce temporal light modulation (TLM)—commonly referred to as “flicker”—that may be visible, annoying, or even debilitating to some, but undetectable to others.

Research efforts into flicker at Pacific Northwest National Laboratory (PNNL) are funded by the Department of Energy’s Solid-State Lighting program and led by Naomi Miller, sometimes known as the “Queen of Flicker” for her work to expand the scientific understanding of TLM and develop metrics and technical guidance to help fix the flicker problem.

"Reports of negative effects from LED flicker by those interacting with lighting are serious enough to warrant research attention,” said Miller. “Understanding flicker and which waveforms have health consequences is the only way we can develop solutions.”

Miller and her PNNL colleagues are exploring the kinds of flicker waveforms that the eye and brain can detect, seeking to understand the different visual and non-visual effects that result. For example, higher frequencies can produce a visual phenomenon called the phantom array effect, where eye movement leads to lights or objects appearing as multiple, separate images like dots or dashes. These phenomena can disrupt and slow visual responses, like reading or driving.

In 2022, the PNNL team published “Flicker: A Review of Temporal Light Modulation Stimulus, Responses, and Measures” in the journal Lighting Research & Technology. Authored by Jianchuan Tan, Felipe Leon, Lia Irvin, and Miller, the literature review examined research to identify what was known and what gaps existed around LEDs and TLM. For example, the review found little research regarding the phantom array effect and no metrics to predict it.

The team conducted an experiment involving 36 human subjects to observe and rate the visibility of both the stroboscopic and phantom array effects under combinations of wave shape, frequency, modulation depth, and duty cycle. Unlike previous work, testing also looked at the differences in response between more sensitive and less sensitive subjects. Results of this study, “Phantom Array and Stroboscopic Effect Visibility under Combinations of TLM Parameters,” were also published in Lighting Research & Technology. The paper, authored by Miller along with Eduardo Rodriguez-Feo Bermudez, Irvin, and Tan, helped differentiate the two responses, and it became clear that a metric for one could not explain the other. The team observed a wide range of sensitivity to the stimuli, although the most sensitive people (those reporting migraine events in the past) were excluded as subjects. This meant the “average” visibility was likely already skewed to the less sensitive side.

This work helped define the frequencies that are visible and thus may cause problems in both average and sensitive populations, laying a foundation for a phantom array effect metric. It was also a first step in guiding electronic driver and dimmer designers toward electronic circuits that minimize the visibility of TLM in LED products.

In 2024, Miller and her team completed an experiment comparing flicker sensitivity of 25 subjects that suffer from migraines and 30 non-migraineur subjects. There was a wide range of phantom array visibility to surprisingly high frequencies of modulation, as well as differences in sensitivity and in post-experiment outcomes between the groups, with migraineurs being significantly more sensitive. A journal article summarizing these results is forthcoming.

TLM is nothing new and was prevalent back when fluorescent lighting relied on magnetic ballasts. In the 1990s, these were widely retrofitted with high-frequency electronic ballasts that nearly eliminated complaints of flicker. The rapid changeover to LED lighting in the last 10 years, however, has made TLM more prevalent and serious. Many LED products do not result in visible flicker. But depending on the TLM waveform and observer sensitivity, the LED product can cause physical responses ranging from simple annoyance and distraction to more serious reactions, which can include headaches, migraines, nausea, and seizures.

Miller notes that PNNL’s work “matters for architects, engineers, designers, and manufacturers who want to feel confident that when they specify or supply architectural lighting, no one using the spaces will be excluded because the lighting makes them ill or uncomfortable.”