Comparing the light quality of retrofit LED products

The rapid development in flux and efficiency of LEDs has resulted in a flooding of the lighting market with solid-state lighting (SSL) products. These are regarded as advantageous alternatives—with superior quality and energy efficiency—to traditional light sources. We found, however, that there are large variations in the quality of these products, and some are not better than the ones they are supposed to replace.

To help consumers make informed decisions and avoid products with unwanted performance characteristics, we conducted a two-year study from January 2010 to January 2012 to investigate SSL products on the Danish market.¹ The focus was on SSL products for replacement of incandescent lamps and halogen spotlights. We collected 266 SSL replacement lamps from lighting companies in Denmark and random samples from stores in the Copenhagen area. Our sample consisted of 144 230V AC lamps (95 directional, i.e., with the luminous flux emitted within an angle of less than 120°, and 49 non-directional) and 122 12V DC lamps (99 directional and 23 non-directional). We tested each light source once for luminous flux, power consumption, and spectral light quality—characteristics essential to long-term satisfaction and therefore technology adoption by consumers—with an integrating sphere spectrometer, using standard measurement guidelines.² We then selected 48 products for a long-term maintenance test, with several measurements conducted at approximately 1000-hour intervals.

The European Union LED Quality Charter (EU-QC) is a voluntary program to promote energy savings in the residential lighting sector.³ We used EU-QC requirements as a reference for evaluation of the light sources in our survey. The efficacy criteria for inclusion under the program increase in steps every year from 2011 to 2015, depending on the color rendering index (CRI)⁴ of the light source. The measurements of these quantities are summarized in Figure 1, which shows the luminous efficacy (a measure of the energy efficiency) of each product as a function of the time of measurement in the study and the EU-QC efficacy requirements. To be included in the charter, the light source must have an efficacy above the line of the same color code.

Figure 1. Luminous efficacy as a function of the time of the first measurement, for the tested directional and non-directional light sources. The colored lines show the requirement for inclusion in the European Union Quality charter (EU-QC). The black lines show a backward extrapolation of the EU-QC requirements. CRI: Color rendering index.

We found that the majority of products tested had a CRI above 80 in accordance with both the EU-QC and widely applied regulation of workplaces.⁵ On the other hand, with regard to efficacy, there is a large spread in the values for similar light quality. Therefore, while most lamps have a high-enough CRI, many fail the efficacy criterion: overall, we found that only 48 of the 194 directional light sources and six out of 72 non-directional lamps satisfy EU-QC requirements.

For the 48 LED products tested in the long-term maintenance procedure, we measured the luminous flux, power consumption, and spectral light quality in the integrating sphere spectrometer at 500, 1000, 2000, 3000, 4000, 5000, 6000, and 11,000 hours of powered operation. Three of the light sources failed completely within the 3000–7500 hour time span. Further, we saw a depreciation of the luminous flux over time for the light sources, evident after 11,000 hours with values from 97 to 80%: see Figure 2. We anticipated this outcome, even though this means some of these light sources might not reach the useful life (typically defined as the time it takes the lamp to reach 70% of initial light output) of 50,000 hours, which is widely quoted. In fact, the depreciation had a large degree of variation from product to product, with some lamps having a useful life of less than 10,000 hours, which is in the range of standard compact fluorescent lamps.

Figure 2. The normalized luminous flux as a function of time for four series of directional LED lamps, each marked with a different color. The significant drop in one of the read series is a sign of a catastrophic failure of the lamp, most likely from electric-driver failure.

Figure 3. Overview of the proposed website (in Danish) showing (a) a list of generic light sources with filters for socket type, directionality, and equivalent incandescent power usage, and (b) individual product information and measurement results for each product.

Our study supports the notion that the market for consumer-grade lighting products has a high variation in efficacy and light quality, with many products not meeting EU-QC standards. Without proper information, it is possible that consumers will find SSL products do not meet their expectations. As a part of our study, we developed a website⁶ for Danes to access some of the information we gathered: see Figure 3. The purpose is to allow consumers to search for the optimal LED replacement for domestically used light sources. A test version of the website was launched in January, and we hope to be able to update the information contained in it with newer products. In the future, we plan to continue the long-term study, with longer intervals between measurements, to test available forecasting methods against measurement.

We acknowledge the financial support of ELFORSK: public service obligation project 342-035, under the Danish Energy Association.