Pre-Conference Tutorials
The two-day workshop began on July 13 but was preceded by an optional half-day of free, beginner-level tutorials on July 12. These tutorials covered the basics on LED lighting—from technology fundamentals; to real-world experiences with LED lighting installations; to results from product testing; to understanding LM-79, LM-80, and TM-21; to using the Lighting Facts® program; to the status of OLED technology.
Competing with Incumbent Technology
A panel discussion entitled "Commercial Ambient Lighting: Can LEDs Compete?" followed, moderated by Marc Ledbetter of Pacific Northwest National Laboratory (PNNL). He set the stage by noting that 2'x2' and 2'x4' troffers with fluorescent lamps dominate the U.S. commercial market and that, because of their high performance and efficiency, those lamps are stiff competition for the profusion of LED products that are intended to replace them but don't yet measure up. Ledbetter observed that, while most of the focus for this market has been on retrofits, as LED efficiency has improved and cost has come down, SSL manufacturers have started coming out with integral replacement products, which compare much more favorably with linear fluorescent lamps. Instead of merely sticking LEDs into a troffer, these manufacturers have rethought the entire design of their integral luminaires, in order to take better advantage of SSL's full potential.
Eric Richman of PNNL discussed a GATEWAY study of LED replacements for linear fluorescent lamps, in which fluorescent T8s and T12s were compared with selected high-performing LED replacement products. He noted that although many of the T8 replacements have good efficacy and their lumen output is increasing overall, they still can't match the light output of their fluorescent counterparts, and their cost is still comparatively high, which is why none of the LED products tested would have paid back within their 50,000-hour reported lifetime. Richman said that for spaces that are currently overlighted, LED replacements may provide an effective solution at correct light levels but may still not make as much economic sense as low-wattage T8 high-performance lamps, which have a lifetime of 36,000 hours.
Gary Trott of Cree LED Lighting Solutions stated that his company's CR24 LED troffer series, which ranges in efficacy from 90 lm/W to 110 lm/W, is priced aggressively and can achieve a payback of as little as two years for new construction. He said improving the design of the product, and cutting its price, required rethinking every aspect of the luminaire—from chip, to optics, to electronics, to thermal management, to mechanical design. Trott added that Cree's CR6 LED series of downlights—an application where SSL has made early inroads because of its directionality—has already reached the inflection point where price is low enough and performance is high enough to cause a rapid increase in sales. "The time is now for LED commercial ambient lighting," he said.
A Rising Tide
Kelly Gordon of PNNL reported on the results of LM-79 photometric testing carried out through the CALiPER test program on a sampling of 33 LED replacement lamps purchased from eight major retailers in the summer of 2010. She noted that large disparities were found between high-performing and low-performing products, as well as wide differences among the retailers themselves, most of whom carried very low-quality products. Gordon observed that most of the LED lamps tested failed to meet basic performance parameters for use as replacements for incandescent or halogen lamps. In addition, differences were found among the manufacturers, some of whose products were very inconsistent across different products. "Obviously, it's in everybody's interest to sell products that will meet the customers' expectations," she said.
Marc Maldoff of Lowe's offered a retailer's perspective on LED replacement bulbs. He described how being a Lighting Facts partner has helped Lowe's hold its suppliers to a higher standard, by requiring them to be Lighting Facts partners themselves, to provide LM-79 data for their SSL products, and to utilize the Lighting Facts label. Maldoff explained that this results in a level playing field for all products, easier product comparisons for retailers, and higher customer satisfaction for consumers. He observed that the best lumen outputs of the LED replacement lamps offered by Lowe's essentially doubled from the summer of 2010 to the summer of 2011, and described his company's efforts to educate consumers on available lighting options and the selection process.
Partners in Promoting Energy-Efficient Lighting
Jon Linn of the Northeast Energy Efficiency Partnerships (NEEP) discussed the DesignLights Consortium (DLC), which was launched by NEEP and has become a major resource for high-quality, energy-efficient commercial lighting design and information. He described how DLC's Qualified Products List for commercial-grade integral LED luminaires came about to help energy efficiency programs distinguish quality energy-efficient lighting products from the rest, and to help them determine whether SSL manufacturer claims are valid, how long the LED product will last, and how comparable it is to its conventional counterpart in terms of light output, color, and light distribution.
Dan Mellinger of Efficiency Vermont, a statewide energy efficiency program, noted that two-thirds of the lighting energy savings they've achieved since 2009 has come from LED screw-base lamps. "LEDs are a huge part of my program right now, and it's come about rapidly," he said. Mellinger discussed the various ways his organization benefits from DOE's SSL program—from CALiPER and GATEWAY reports, to Technology Fact Sheets, to being a Lighting Facts and L Prize partner, to membership in the Municipal Solid-State Street Lighting Consortium and the Technical Information Network for Solid-State Lighting. He recommended that other programs emphasize quality in order to maximize savings and maintain customer satisfaction, and that they educate staff, partners, and customers on the technology and best practices.
Brodrick concluded the panel discussion by giving special recognition to nine Lighting Facts partners who exemplify how integrating the Lighting Facts program into their everyday business practices can help ensure that quality SSL products are available:
- Lowe's, for making Lighting Facts testing procedures a requirement for all SSL suppliers and incorporating LM-79 test data into its product evaluation process
- Efficiency Vermont, for being the first energy efficiency (EE) sponsor partner to register a program on the Lighting Facts EE Partner Resource, and for encouraging partners and customers to look for the Lighting Facts label
- The Sacramento Municipal Utility District, for using Lighting Facts when researching potential products for incentive programs, and for using the label to educate customers on the benefits of LED products
- Acuity Brands, for supporting the use of Lighting Facts for the commercial, industrial, institutional, and residential markets, and for using it to educate retail customers
- BetaLED, for stressing application-based performance evaluations to meet customer expectations, and for using the Lighting Facts label to enhance credibility with customers
- OSRAM Sylvania, for using the Lighting Facts label when introducing new products, and for including the label on all LED retrofit products, allowing consumers to compare information
- Satco, for submitting every LED product for a Lighting Facts label, and for being committed to educating customers about LED technology
- Grainger, for requiring all LED suppliers to take the Lighting Facts pledge, and for incorporating Lighting Facts into its corporate strategy
- The Home Depot, for requiring all suppliers to complete standardized quality testing and to provide a Lighting Facts label, and for being committed to educating the public on LED technology
Brodrick noted, "We'd like everybody to join the Lighting Facts program," and invited attendees to learn more at the evening reception and partner poster session, where attendees had an opportunity to browse posters highlighting various partnership opportunities and ask questions of the various DOE partners.
Lessons from the Real World
Kevin Powell of the U.S. General Services Administration (GSA) discussed his agency's Green Proving Ground program, which tests and evaluates innovative energy-saving technologies and, to date, has done 72 projects that have incorporated some form of LED lighting. He pointed out that lighting accounts for nearly one-third of the energy use of commercial buildings, and talked about "smart" lighting's considerable promise, noting that it can facilitate energy savings through the use of controls and make maintenance easier by reporting lights that are malfunctioning. But Powell also talked about the challenges smart lighting faces, such as overcoming a high first cost. He said that lighting innovation is the key to meeting GSA's mandate of zero net energy by 2020, and that reducing the agency's lighting energy consumption by 60 percent to 80 percent by then is a realistic goal.
Edward Smalley of Seattle City Light, who directs DOE's Municipal Solid-State Street Lighting Consortium, described his city's progress in its plan to replace 41,000 residential street lights with LED fixtures. In addition to actual energy savings of 48 percent, he cited maintenance savings and improved safety from better lighting quality as among the driving forces behind the project, which began in 2007. Smalley discussed the luminaire selection process he uses, and noted that SSL costs are constantly coming down, making the technology more and more affordable. He added that the use of controls, which Seattle is considering, has the potential to save even more energy. "LED street lighting has proven to be a significantly better light source in terms of expected maintenance, energy efficiency, and quality of light," he said.
Defining "Quality of Light"
Next, a panel moderated by Eric Richman explored the issue of color quality. Maria Thompson of OSRAM Sylvania discussed defining the quality of light. She explained that color quality has two main aspects: chromaticity, which is the color of white light; and color rendering, which is the color of an object as illuminated by a light source. Thompson said that although chromaticity is mainly measured by CCT, CCT by itself can't communicate chromaticity when it's outside of the black-body curve, which is why Duv, which measures distance from the black-body curve, is also used. Thompson also delved into the subjects of color fidelity and color saturation, noting that while a high CRI means good fidelity, it doesn't necessarily mean good color saturation, because saturated colors aren't part of the CRI average. She said the Color Quality Scale, recently developed by the National Institute of Standards and Technology, partly addresses this weakness of the CRI.
Daniel Salinas of Nelson Electric approached the topic of defining the quality of light from the point of view of a lighting designer and an electrical contractor, as he wears both hats professionally. He said that if LED luminaires from various manufacturers are being used on the same project, color uniformity is a critical consideration, and differing control requirements can cause installation conflicts and confusion. Salinas suggested using the same manufacturer's product for similar applications, to minimize any variance in LED color or distribution. He also suggested establishing the acceptable amount of color shift for the project in question and making it part of the binning requirements. Salinas cited accessibility as a major installation consideration, and stressed the importance of ensuring adequate ventilation.
Emphasizing that the quality of light is "more complex than just lumens," Chad Stalker of Philips Lumileds discussed the various performance drivers for LED quality of light. "With light, there are human, economic, and architectural factors," he said. Stalker observed that LEDs can support the broad color spectrum needed for different applications, and noted that natural colors render best under sources that mimic natural light. He remarked that staying within three MacAdam ellipses ensures color consistency between light sources, because the human eye can't detect changes in color that are less than that. Stalker advocated testing and binning LEDs at operating conditions in order to eliminate unknowns, simplify the design process, and lower costs.
In the Driver's Seat
Matthew Reynolds of National Semiconductor said innovation—especially with the drive and control of the LED light source—is key to driving the evolution of the still-emerging SSL market. He remarked that although the current focus is on using LEDs to replace and retrofit existing lighting installations, this focus will shift to developing intelligent lighting systems in order to fully realize system efficiency savings. Following that shift, Reynolds added, will be an emphasis on leveraging the controllability of LEDs—for example, tuning the color and controlling the color temperature to optimize the lighting in a particular space. He said the electronic drive and controls will be the keys to unlocking and leveraging the benefits LED light sources offer over all other light sources. "We want to create an SSL system that's as good as, or better than, the incumbent system," Reynolds said.
Julian Zhu of NXP Semiconductors focused on dimming—specifically line voltage dimming, which is the most popular dimming method. He observed that when using line voltage dimmers, it's more difficult to dim an LED lamp or luminaire than an incandescent lamp. Zhu said line voltage dimmers add design complexity to LED drivers, which can cause the lamp or luminaire to flicker if they're not designed properly. He discussed a number of specific dimming challenges; for example, in order to support the reliable, flicker-free operation of Triac dimmers, the LED driver has to provide the latching and hold current when the system current is not enough, but this adds bleeding loss to the driver, thus making it more challenging to achieve high efficiency.
Pantas Sutardja of Marvell Semiconductor discussed design for high-quality and low-cost solid-state lighting with power factor correction. He said that lifetime output control increases the cost slightly, because the feedback sensor design needs to be robust against shading and reflection, but that most of the other costs can be absorbed into the LED driver design. These other costs include the final driver controlled by output voltage and power handling spec, as well as sophisticated features provided by control and sensing circuitry. Sutardja said that high CRI and the desired CCT can be achieved by dual string color mixing, and well-behaved dimming can be made available at almost no extra cost, but driver manufacturers need high sales volume to make this economically feasible. "We're really close to having all these things available in the near-term," he said.
The Question of Lifetime
Terry Clark of Finelite, a member of that same working group, focused on the issue of color shift over time. He explained that although it wasn't part of the standard lifetime definition the group recommended, it's of paramount importance for some applications. Clark reviewed the known causes of color shift in luminaires and noted that it's a characteristic of most light engines, including metal halide, incandescent, and fluorescents. What makes it so critical an issue, he said, is that its causes with LEDs aren't well-understood, and testing shows that color shift in LED-based luminaires can be so pronounced as to constitute failure to an end user. Clark said a lot more work needs to be done on the issue, which is a complex one. "We don't have a great vocabulary or common, shared way to describe color, and describing the shift over time is really a challenge," he said.
Stressing the importance of reliable design, Steve Paolini of Lunera Lighting reviewed the various components of a luminaire that can affect its lifetime. He said that in a well-designed LED luminaire, the LED current is protected from electrical mains transients, and the LED junction and components are kept below the maximum tolerable temperature. Paolini observed that the demand for long life is partly driven by SSL's high initial cost, but noted that long life increases the price because it entails more costly components, requires testing, and is backed up by a warranty. He emphasized that the lifetime required of an SSL product depends on the application, with some applications—such as those where maintenance is difficult and costly (e.g., bridge lighting)—needing longer lifetimes than others. "Having a shorter-life product may be a good tradeoff," he said.
SSL Product Pricing
Mike Watson of Cree offered a manufacturer's perspective on cost-reduction strategies. He noted that LED lamp costs have dropped dramatically over the past five years, and the trend should continue. But Watson said the current focus on LED cost is misplaced, because improving LED lamp performance will yield greater cost reduction than optimizing manufacturing or individual LED lamp price, since it will allow for the use of fewer LEDs per package. As an example, he described how Cree took its first-generation 650-lumen, 12W LR6 downlight, which had 42 LEDs and wholesaled for more than $100 when it appeared in 2007, and engineered the costs down to produce the second-generation 575-lumen, 10.5W CR6, which came out in 2010, has eight LEDs, and retails for $50.
Mark Hand of Acuity Brands Lighting talked about how the correct specification can reduce the cost of SSL. He raised the question of whether customers expect longer life, better color quality and distribution, less energy consumption, and more control from LED luminaires because of their high price—or whether their price is high because customers expect such high performance. Emphasizing that many SSL systems are over-specified for the application, Hand examined a number of lighting parameters to show how they each allow considerable leeway, depending on the application. For example, while 50,000-hour lifetimes may be suitable for applications where the lights are operating 24/7, shorter lifetimes may make more sense for less-intensive applications and can lower cost by allowing the LEDs to run hotter, which can result in more lumens and fewer LEDs.
Brodrick concluded the workshop by thanking participants for their input and participation. He noted that the next DOE SSL workshop is in February 2012—the ninth annual Solid-State Lighting R&D Workshop—and also encouraged attendees to stay apprised of DOE SSL program activities by visiting www.ssl.energy.gov.
Workshop attendees got a chance to see some of Seattle's LED street lighting installations firsthand on a guided bus tour the last night, which served to unofficially kick off the Municipal Solid-State Street Lighting Consortium's one-day regional workshop that was held the following day. Hosted by Seattle City Light, the bus tour was led by Edward Smalley, who shared lessons learned along the way. Attendees visited a number of residential and arterial LED streetlight sites throughout the city, participated in evaluating them, and in some instances were able to compare the LED streetlights with the incumbent HPS street lighting.