You are here

In 2017, the U.S. Department of Energy began evaluating the performance of several new light-emitting diode (LED) lighting products with claimed efficacies near or exceeding 200 lumens per Watt (lm/W). The purpose of the study was to validate the claimed efficacy, determine the features or characteristics the products had in common, and identify any performance tradeoffs in luminaires reaching this very high efficacy level.

Two samples each of seven luminaires – five identified through their LED Lighting Facts listings, and two found through an online search – were anonymously ordered. All of the products were industrial luminaires with correlated color temperatures (CCTs) of 5000K, and all but one had exposed LED emitters. The results of photometric testing of the samples were compared with the manufacturer-claimed values. Additionally, samples were tested for horizontal illuminance, flicker, and maximum luminance, and were also evaluated by 23 knowledgeable observers.

Two photos showing a mockup space with luminaires installed, and laboratory sphere testing of a high-efficacy luminaire.
Mockup space with luminaires installed (left). Laboratory sphere testing of a high-efficacy luminaire (right). Photos courtesy of Pacific Northwest National Laboratory.


Overall, photometric testing of the seven luminaire pairs yielded values that were reasonably close to the manufacturers’ claims, with actual lumen output varying from the claims by no more than 9.6%, and actual power draw by no more than 6.8%, while power factors were all above 90%. Efficacies varied from manufacturer claims by as much as 12%, although the tested values were often considerably lower than those listed in LED Lighting Facts, which are based on the top performers in a product family. Actual CCT and color rendering index (CRI) were consistent with claimed values.

One of the seven luminaire types exhibited flicker exceeding the low-risk criteria of the Institute of Electrical and Electronics Engineers Standard PAR1789-2015. When installed in a high-bay space, all of the luminaires produced a fairly even light distribution across the workplane, with a maximum-to-minimum illuminance ratio below 1.7 at a 37-inch standing desk height. Direct luminance measurements of the exposed LEDs ranged from 154,000 candelas per square meter (cd/m2) up to 478,000 cd/m2, although due to the difficulty of measuring LED luminance directly, these are likely lower than the actual luminances. The one luminaire that used diffusing tubes to cover the LED emitters had a luminance of 40,000 cd/m2 (for comparison, a T5 high-output fluorescent lamp has a luminance of about 25,000 cd/m2).


In light of the high maximum luminances, it’s not surprising that glare was the greatest complaint from the observers. Only two of the luminaires received acceptable ratings for visual comfort and overall quality, and those had either diffusing lenses or reflector optics engineered to cut off the view of the bare LEDs above a fixed viewing angle. The two top-rated luminaires received acceptable ratings in categories of light distribution, shadows, and color, in addition to visual comfort. The least-preferred luminaires corresponded to the three that received the most negative comments about glare.

The fact that the product with the highest rating also had the lowest tested efficacy (136 lm/W) shows that visual comfort should also be considered. There are always tradeoffs in selecting the best luminaire for an application, and efficacy is only one performance aspect to consider. Other attributes – such as visual comfort, light distribution, flicker, shadows, and color quality – are all important.


It’s important to understand that the efficacy numbers listed in a product database such as LED Lighting Facts represent a particular model and set of characteristics across a family of products. Significant variation within a product family is common, due to differences in lumen output (the highest-efficacy product is often the version with highest output), light distribution (efficacy is reduced by optics), and color temperature (the highest efficacies are generally associated with the highest CCTs; e.g., among the seven products tested, dropping from 5000K to 3000K resulted in as much as a 17% loss in efficacy). Efficacy and other individual attributes should only be used to identify products to consider and should be supplemented by the collection of more detailed performance data. Whenever possible, specifiers and purchasers should see the actual product in action before making a selection.

The study shows that there are LED lighting products with very high levels of efficacy. While databases such as LED Lighting Facts can be excellent resources, they don’t necessarily include products from every manufacturer. Additionally, specifiers and purchasers should investigate the performance of the specific product needed – e.g., by gathering additional information from the manufacturer’s website, since the listing data may be based on the performance of a product very different from the one that is needed. With due diligence, high-performing LED products can be found for virtually all architectural applications – though tradeoffs will usually need to be made to address such lighting-quality issues as comfort, color, and distribution.