If you are interested in how solid-state lighting performs when used in recessed downlights, track lights, A-lamps, cove lights, and replacement lamps for linear fluorescents, you might take a look at the Round 12 results recently released for the Dept. of Energy's CALiPER (Commercially Available LED Product Evaluation and Reporting) program.
One conclusion: LED light is getting warmer. Says the report, "The majority of the SSL products tested have rated CCT (correlated color temperature) of 2700 or 3000K.... Showing clear progress over earlier rounds of testing, only two SSL products have CCT ... outside standardized tolerances for white light in SSL products. Color rendering characteristics are also improving with an average CRI (color rendering index) of 83 and only one product with CRI below 75."
And utilities will probably be happy to know that the average power factor for the fixtures DoE tested was 0.9. Moreover, SSLs are getting more efficient, says DoE. "The steady increase in average and maximum efficacy is clear. The minimum efficacy seen in Round 11 is actually higher than the overall average efficacy observed in 2007 (26 lm/W minimum Round 11 versus 21 lm/W average in 2007)," says the report.
But average efficacy actually declined slightly in the Round 12 results, to 46 lm/W. Nevertheless, that figure actually exceeds the average efficacy for warm-white SSL products tested in previous rounds. In those tests, many of LED products tended to fall into the cool-white range. So the efficacy trend is still rising when comparisons are between LED products with comparable color.
At least from the standpoint of efficiency, SSL came out of the tests looking good. In downlights, for example, two of the ten Round-12 SSL products exceed the highest performing six-inch downlight products checked in previous tests. The Round 12 SSL recessed downlights were better than recessed downlights equipped either with incandescent A-lamps or with Halogen HIR PAR38 bulbs, achieving three to six times the luminaire efficacy of incandescent and halogen, says the DoE. And all ten SSL recessed downlights exhibited better efficacy than twosix-inch recessed downlights equipped with 26-W and 32-W triple-tube, pin-based CFL lamps.
But DoE found some problems with light labeling. Six of eight SSL A-lamps it tested carry the Lighting Facts label. (A lamps are SSLs that mimic an ordinary incandescent bulb envelope and screw into a light socket.) All six met manufacturer ratings, but one was borderline (tested at about 10% below values claimed on the Lighting Facts label). Similarly, five of the six lamps with Lighting Facts labels provide accurate equivalency statements, but one lamp (which is borderline on meeting ratings) claimed equivalency to 25–60W incandescent, but meets the light output of only a 35-W incandescent. Both lamps lacking Lighting Facts labels had inflated equivalency claims, and one didn't meet manufacturer-rated performance levels.
Other results pertained to track lights, cove lights, and several other categories. The full report is available free and can be found here: http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/ns/caliper_round12_summary.pdf