New study provides evidence for more stringent ”flicker” requirements in pending ecodesign lighting regulations

(eceee news, 14 Dec 2018) Just days ahead of an important EU vote on lighting ecodesign regulations, a new study provides evidence that significantly more stringent requirements on “flicker” could be justified to protect health and well-being. Other data indicates that the effect on price of lighting sources would be negligible. Over a billion light sources are expected to be sold in Europe before the regulation can be revised.

The study is carried out by researchers at the National Research Council of Canada and the Centre Scientifique et Technique du Bâtiment in France, two of the world’s leading research institutions on lighting and health. The IEA 4E Solid State Lighting Annex and six of its member countries have commissioned the study.

“Flicker” in light sources is a popular word for “Temporal Light Modulation”, often abbreviated as TLM. Temporal variation in light output from a light source is known to, under certain circumstances, have an adverse impact on people’s health. Recognised effects range from the distraction of visible flicker, health ailments such as eyestrain, and reduced task performance, through to critical health conditions such as migraines, photosensitive seizures and autistic behaviour.

The new ecodesign “omnibus” lighting regulation will be voted on by the Member States in the Ecodesign Regulatory Committee meeting on Monday 17 December. The Swedish government, one of the funding bodies of the study, has submitted a copy of this study to the regulatory committee members which eceee has obtained.

Although Sweden did not commit to a specific level at the time they submitted the study, the information published provides a very strong foundation for more stringent TLM requirements. The proposed level in the current proposal is slightly higher than the levels observed for old magnetic- ballasted T12s in the 1980’s, which were found to cause headaches and eyestrain. Considering that over a billion long-life LED light sources are expected to be sold in Europe before the regulation will be updated, the outcome of this decision is significant. And research on the incremental manufacturing costs associated with flicker-free LED driver designs indicate that more stringent TLM requirements will have no effect on consumer prices (as is the case in today’s market).

Considering the large number of consumer to be affected – the EU alone has over 500 million inhabitants – reducing the percentage of the population potentially affected, would translate into many millions of people being affected or not. Even one percent of people affected translates into 5 million European citizens.

What would be relevant levels?

LED light sources are basically electronic components with very fast response times and are controllable in ways not experienced with traditional light source technologies. The draft Ecodesign measure for lighting sets requirements on two metrics – PstLM, the short-term flicker metric for visible flicker and SVM, Stroboscopic Visibility Measure. Taken together, these two metrics cover the most important flicker frequencies for people.

The draft regulation sets limits of: PstLM ≤ 1.0 and SVM ≤ 1.6, however these have drawn criticism from health authorities as being too high. For this reason, six governments who are all members of the IEA 4E SSL Annex conceived and launched a subject-based study focusing on SVM and the threshold detection level.

The report presents the results of an experiment designed to find the level of detection at varying levels of SVM in a sample of the population. The interim findings of this study indicate that specifying a maximum level of 1.6 will not address the issue of SVM in LED light sources. The report shows that:

  • The most sensitive 25% of the people detected stroboscopic effects in the experiment 90% or more of the time at a lower level of SVM=1.4 (75th percentile overall), suggesting that the maximum level for SVM must be lower.
  • The experiment found that the most sensitive 25% of the sample could still detect the stroboscopic effect more than 63% of the time at SVM 0.9. Thus, to protect this group, the maximum SVM level should be lower.
  • The greatest certainty would be provided by setting a level closer to SVM 0.4 as at this level and below, the detection rate for the most sensitive quartile of the people detecting the stroboscopic effect dropped to 10%.

The proposed level of 1.6 SVM was already known to be slightly higher than the levels set for old magnetic- ballasted T12s in the 1980’s, which were known to cause headaches and eyestrain.

The Swedish Energy Agency’s cover letter submitted to the regulatory committee can be found here.

The study itself can be found here.