Tang Yau Hoong

In response to “LEDs Make Inroads into Streetlighting” (Jan/Feb 2015), Jim Benya urges the lighting community to focus its attention on LED lighting quality, lest a dangerous precedent be set that ignores color temperature and makes way for glare in the name of energy efficiency.


For me, LED lighting has now surpassed all expectations. For years, I wondered whether manufacturers would develop sources and luminaires with enough power for commercial lighting and, to my amazement, those now exist. Even sports lighting, one of the most demanding applications in terms of both controlled candlepower and raw lumens, is now possible. As a result, I am finally telling my clients that all-LED lighting represents the best lighting investment they can make, even if it costs a little more than legacy systems.

For good reason, LED lighting has been heavily promoted for both its energy efficiency and long life, as well as for convenient dimming, color temperature tuning, and more. But one rarely hears of LED products touted for their lighting quality. To the contrary, I think we’ve been far too lenient about quality deficiencies that many LED products introduce.

In my opinion, lighting quality begins with white light source color. With solid-state lighting, there are only three practical ways to make white light: phosphor with a blue pump, phosphor with a violet pump, or a combination of narrow-band LED sources (RGB, RGBA, etc.). The most common are blue pumped, but any of these are capable of producing white light. As a general rule, the quality of the white light is inversely proportional to the efficacy (lumens per watt) of the source. The difference in efficacy between an 80 CRI LED and a premium 95-plus CRI LED can be 25 percent to 30 percent; between a 65 CRI LED and a 95-plus CRI LED, a whopping 40 percent to 50 percent.

When efficacy wins, color quality suffers. The highest-efficacy white LEDs produce little to no red, little to no cyan, an abundance of green, and an over-abundance of blue. Blue worries me most for environmental reasons, because the presence of abundant blue signals daytime to the circadian system. Other color defects, such as unacceptably poor color rendering of critical tasks is also of concern.

After selecting the source, I look for the ability to control the light. The vast majority of LED devices emit a semi-Lambertian distribution—what we technically call a blob. A lot of simple luminaires add nothing more, resulting in a fixture that throws light everywhere in front of it. Not surprisingly, these luminaires produce the most lumens per watt. For proper lighting design practice, this is generally unacceptable, as specific candlepower and beam shape is needed to do almost everything right. Unfortunately, beam quality is not a criterion on the DesignLights Consortium Qualified Products List. Efficiency and low cost will appeal to far too many, I’m afraid.

The third issue I then focus on is flicker. During the magnetic ballast era, we learned that flicker associated with fluorescent and high-intensity discharge lighting was bad. It caused headaches and its stroboscopic effects could be dangerous in an industrial workplace. Why, all of a sudden, is blatant flicker now casually ignored? Have humans evolved, or is it too hard to eradicate it from LED lighting?

But the biggest potential problem with LEDs, by far, is glare. The on-axis luminance of a bare, modern, high-power LED exceeds 100 million cd/m2 (nits), and each year products get brighter, so we can expect higher values in the future. The luminance of the sun at noon is 1.6 billion nits, making bare LEDs now about 1/16 of the brightness of the sun (and getting brighter). LEDs are typically arranged into a grid in order to reach the total lumen output desired. By any glare formula in the world, their glare vastly exceeds the maximum acceptable luminance of any light source, for any reason.

We can solve this problem of glare with a combination of shielding, refraction, diffusion, indirect lighting, and illuminating adjacent surfaces to reduce contrast. Quality luminaires do some or all of these. But the vast majority of LED products worldwide do nothing but produce excessive brightness over a wide field of projection. I’ve taken to describing the effect “pincushion glare” to call attention to the extreme brightness of individual LEDs. Each LED image is focused onto the retina and some individual cones will be overdriven to the point of pain. As an added insult, the excessive blue of many LEDs severely overdrives the blue cones, causing hypoxia, resulting in temporary blindness, long-lasting afterimage, and, for most people, a glare reflex.

What can we do? With the overarching emphasis on energy efficiency on which LEDs feed, suggesting better LED luminaires with fewer lumens per watt may be seen by some as heretical. But, now that the sources have become so efficient, perhaps it is time to rethink the quality compromises made in the name of efficiency, and to stop the race for the bottom before good lighting design practice is the loser.

James R. Benya, PE, FIES, FIALD
Benya Burnett Consultancy, Davis, Calif.