Flicker Studies to Guide Future SSL Systems

A newly published study that tested the effects of flicker on humans offers recommendations that will become the basis for future Alliance for Solid-State Illumination Systems and Technologies (ASSIST) guidelines for the LED and lighting communities.

The lighting industry and the Energy Star program have debated the effects of frequency and other driving modes on the perception and acceptability of flicker, and to provide further data and guidance in this area, the Lighting Research Center (LRC) at Rensselaer Polytechnic Institute conducted the studies funded by ASSIST.

Both photographs were taken under a flickering light source (at 120 Hz) with an exposure time of 1/15 of a second. In the top photograph, the ruler is stationary and no stroboscopic effects are seen. In the other photograph, multiple images are produced by each flicker cycle as the ruler moves across the scene. (Image: Lighting Research Center/RPI)

Because of the numerous ways in which LEDs can be driven, there is a potential for a wide variety of flicker characteristics in solid-state lighting (SSL) systems. These can be perceived directly if the flicker itself is visible, or indirectly through stroboscopic effects similar to the “wagon-wheel effect,” where a spinning wheel appears to be moving slowly or be stationary under intermittent light. In early 2010, industry leaders from ASSIST asked the LRC to update the available research literature on perception of light source flicker through human factors studies. Previous research had shown frequencies at which direct flicker is perceptible, but ASSIST leaders expressed interest in identifying thresholds and acceptance levels for indirect perception of flicker and a means of predicting these levels for SSL and conventional light sources.

In the first published study, led by Dr. John Bullough, LRC senior research scientist, a table lamp in a laboratory was fitted with LED sources to produce various flicker frequencies, modulation depths, duty cycles (duty cycle is defined as the percentage of time a modulating light source is “on”), waveform shapes and correlated color temperatures. Study participants reported whether they detected flicker effects and if so, how acceptable they were. They also rated their visual comfort under each condition.

Bullough and colleagues found that, although flicker was not directly visible at frequencies of 100 Hz or higher, indirect stroboscopic effects of flicker were perceptible even at 300 Hz. Lower modulation depths substantially reduced the perception of stroboscopic effects, and a higher duty cycle resulted in somewhat higher rated comfort than a lower duty cycle. Neither the shape of the flicker waveform nor the correlated color temperature of the light affected responses to flicker under the conditions studied.

“The results suggest that there is a tradeoff between the frequency and the modulation depth in the detection and acceptability of indirect flicker effects,” Bullough said. Building on these results, ASSIST and the LRC performed a follow-up study to systematically evaluate this tradeoff and look more closely at the relationship between frequency and percent flicker. A second paper describing this study has been accepted recently for publication.

With the published findings, ASSIST is developing a set of recommendations for manufacturers under its ASSIST recommends publication series, said Dr. Nadarajah Narendran, LRC director of research and ASSIST organizer. The recommendations will be made available for download later this year on the ASSIST website.

For more information, visit: www.lrc.rpi.edu/assist