Standards Deviant

All lighting products sold in the United States are subject to industry standards governing safety and performance. Because testing and rating systems must be applied in the same way, these national standards enable consumers to compare products made by different manufacturers, or even across technologies, such as incandescent, fluorescent, and high-intensity discharge (HID). However, solid-state lighting (SSL) technology is a very different light source than its predecessors, making it an outcast from the existing standards covering the mechanical forms, electrical connections, and measurement of traditional lighting technologies.

The development of SSL technologies — LEDs and OLEDs — is undergoing rapid change and improvement, taking them from niche applications to possible general illumination. However, as is the case for most burgeoning technologies, the products arriving on the market exhibit a wide range of performance. Driven by marketing, the metrics about these products, especially LEDs, are sometimes hyped with incorrect figures, such as performance reported in lumens per watt without a disclosure of the price, package type, chip size, or measurement technique. As a result, the LED industry has been plagued by dashed expectations, skepticism, and consumer confusion.

Apples to apples

“The reason you need standards is because the application communities interested in using this technology would like to be able to quantify the performances of these systems — so that when they specify something then they know they can be relied upon,” says Nadarajah Narendran, Ph.D., associate professor at the Rensselaer Polytechnic Institute (RPI), Troy, N.Y., the director of research at the Lighting Research Center (LRC) at RPI, and the organizer of the Alliance for Solid-State Illumination Systems and Technologies (ASSIST) program. He is also a fellow of the Illuminating Engineering Society of North America (IESNA), New York. “If there is no standardized way of measuring the technology, then you really don't know whether claims by different manufacturers all mean the same thing. There could be a misunderstanding on how things are being measured and quantified.”

SSL differs fundamentally from traditional lighting technologies in terms of materials, drivers, system architecture, controls, and photometric properties. The average life of a particular lamp type is the point where 50% of the lamps in a representative group have burned out. The life of an LED depends on its packaging configuration, drive current, and operating environment.

“There's a fundamental electrical incompatibility between how we distribute power within a building and what an LED's power needs are, just looking at one single device,” says Rob Arnold, electrical engineer and proprietor of the online LED Center (http://led.linear1.org/) in DeSoto, Kan. “It wants a few volts and a small amount of current, and the distribution network in a house is set up to produce a lot of volts and a lot of current.”

Arnold explains that there needs to be a driver in between the LED and the distribution system. “This is where standardization would really benefit the marketplace,” he says. “If there were a standardized LED driver component that certain standardized modules of light-producing light engines could plug into, that would really benefit the guy that has to specify the solution. The lack of standardization means that every design problem requires a unique solution.”

There is no standard procedure for rating the light output, or luminous flux, of LED technology because it doesn't account for driver losses. LED devices are tested using a pulse measurement at room temperature — to run them longer without a heat sink would damage them — whereas LED fixtures are evaluated using steady-state measurements. Furthermore, there is no standard test procedure for measuring the luminous flux of LED arrays, such as multiple LEDs mounted on a circuit board.

Therefore, luminous flux — and by extension, luminous efficacy — must be measured at the luminaire level. Recently released results from Round 4 of the U.S. Department of Energy's (DOE) Commercially Available LED Product Evaluation and Reporting (CALiPER) Program, conducted from September to December 2007, demonstrated that many commercial LED fixtures don't live up to their manufacturer's stated values of efficacy and lumen output. A common problem is that a fixture datasheet may quote the efficacy value for an individual LED. Within the fixture itself, depending on the quality of its design, optical and electrical losses can easily reduce the efficacy by 30%. The new Energy Star ratings for LED fixtures are all based on luminaire — rather than LED or LED system — efficacy.

The color-rendering index (CRI), which indicates how well a light source renders the colors of objects, materials, and skin tones, has been used to compare fluorescent and HID lamps for more than 40 years. However, the International Commission on Illumination (CIE), Wien, Austria, which published the CIE 13.3-1995, “Method of Measuring and Specifying Colour Rendering Properties of Light Sources,” does not recommend its use with white light LEDs. “The CIE CRI is generally not applicable to predict the color rendering rank order of a set of light sources when white LED sources are involved in this set,” states the conclusion of the technical committee in its CIE Technical Report 177:2007, “Color Rendering of White LED Light Sources.”

Instead, a long-term research and development (R&D) process is underway to develop a revised color quality metric that would be applicable to all white light sources. In the meantime, DOE recommends that CRI only be used as one data point in evaluating white LED products and systems — not to make product selections in the absence of in-person and onsite evaluations, particularly in spaces where color or fabric comparisons are made under daylight and electric lighting.

Measure for measure

Industry groups, standards setting organizations, and the solid-state lighting division of the DOE's Energy Efficiency and Renewable Energy Building Technologies Program are moving quickly to develop needed standards and test procedures for SSL products. In the meantime, however, there is a need for reliable, unbiased product performance information. In the absence of standards, the DOE offers the results from its Gateway Demonstrations, which showcase high-performance LED products for general illumination in a variety of commercial and residential applications. Demonstration results provide real-life experience and data on SSL product performance and cost effectiveness. These results connect DOE technology procurement efforts with large-volume purchasers and provide buyers with reliable data on product performance.

“We photometrically test the products and match up the ones that work well with a host site,” says Jim Brodrick, SSL portfolio manager, DOE. “The host site gets the lighting for free, and we conduct a survey at the site. We get experience and hard data about how these things really operate. The photometric testing is done by DOE, not the manufacturer, so the data that comes out is the way it is.”

Test results from the DOE's CALiPER Program — formerly the SSL Commercial Product Testing Program — help guide DOE planning for future SSL R&D and commercialization support activities, support DOE technology procurement activities and associated technology demonstrations, provide objective product performance information to the public in the early years, and guide the development, refinement, and adoption of credible, standardized test procedures and measurements for SSL products.

Through the program, DOE supports testing of a wide representative array of SSL products available for general illumination, using test procedures currently under development by standards organizations. Guidelines for selecting products for testing will ensure that the overall set of tests provides insight on a range of lighting applications and product categories, a range of performance characteristics, a mix of manufacturers, a variety of LED devices, and variations in geometric configurations that may affect testing and performance.

In results from Round 4, DOE found that while some products tested in this round showed progress in performance compared to products tested a year ago, half of the products tested in this round are still achieving efficacy levels that are only slightly more than would be expected in similar products using halogen sources. Similarly, when considering appropriate light output levels and color characteristics for each application, about half of the products tested in Round 4 would be suitable substitutes for products using other sources, while half would have levels of output that might be considered too low or color qualities that might be inappropriate as compared to products using traditional sources for the same application.

This study also found that not only were efficacy levels in half the products not up to snuff, but that there were several discrepancies between the light outputs and efficacies published by manufacturers and their CALiPER-tested performance. Out of 15 SSL products tested in this round, one manufacturer provided accurate performance information for its luminaire, four products offered no manufacturer-published information regarding output or efficacy, one product manufacturer's literature understated the output and efficacy of its SSL product by 50, and nine products contained information published by manufacturers that overstated performance reporting by factors ranging from 30% to 600%.

Of these nine manufacturers, some compared the outputs of their products to incandescent products; some have published explicit but inaccurate values for the output or efficacy of their luminaire; and some published values for output or efficacy, but didn't indicate what the values correspond to. The report credits the overstatements to misinterpretation or lack of experience relative to SSL testing concepts, lack of industry standardization in LED device performance testing and reporting, and infeasibility of determining luminaire performance based on reported LED device performance. In addition, overstatements may also be caused by confusion or lack of clear distinction in marketing literature between LED device performance and luminaire performance, use of inconsistent testing methods that may yield different results, a lack of specificity to indicate what specific product configuration was tested to produce the performance values published, or possible inflation of performance claims or selection of test conditions not representative of actual use.

“Addressing and resolving these issues should be of vital concern to SSL manufacturers,” reads the report. “Performing appropriate SSL testing and providing accurate, understandable information regarding product performance will increase confidence in SSL technology. Continuing to provide inaccurate or misleading SSL product performance information may undermine market acceptance of this new technology in the long term.”

Setting standards

Although misleading or confusing specifications might not be a deal breaker for lighting professionals who can sort through manufacturer claims to find the real points of comparison, they might hinder LEDs from achieving a broader market penetration. “There are adequate specifications and details in place that a lighting designer or an architect familiar with the technology could specify a solution and work with a supplier to produce an installation that meets whatever the design specs are,” Arnold says. “It's not an unsolvable problem, but this won't work to get mass market penetration.”

Arnold compares LED technology to the marketing of compact fluorescent lighting (CFL), which over the past few years has taken the market by storm. As Arnold points out, a CFL is marketed in terms of the equivalent wattage of the incandescent lamp it's intended to replace. When people purchase a CFL, they buy the one that's intended to replace their former incandescent lamp so the CFL will be labeled with the wattage of its predecessor, 100W (for example), even though the device's actual power consumption is less. “In terms of solid-state lighting, it's a challenge bringing these things to market when we don't have a way to specify how much light we get out of them,” Arnold says. “The ways of measuring light are not necessarily highly accessible to ordinary consumers.”

To accommodate the characteristics of LED technology, some existing standards and test procedures are being modified while, in other cases, new standards are under development (see Product Performance, Measurement, and Safety Standards on page 29). To accelerate the development of needed standards for SSL products, DOE facilitates ongoing dialog and collaboration with key standards-setting organizations, and offers technical assistance in the development of new standards.

“We're working with the six or seven standards-writing bodies in the nation, and we are basically facilitating them to get these needed standards done sooner than later,” Brodrick says. “Sometimes we help out just by convening meetings and making communication quicker and faster, but we also run a certain amount of tests so they have hard data with which to make decisions.”

Rosslyn, Va.-based National Electrical Manufacturing Association's (NEMA) SSL Section was founded to address the issue of standards, as was the Washington, D.C.-based American National Standards Institute (ANSI) Working Group C78-09 and the IESNA Testing Procedures Committee (TPC) SSL subcommittee. During the first quarter of 2007, Northbrook, Ill.-based Underwriters Laboratories (UL) formed a balanced Standards Technical Panel (STP) to work on the drafting and publishing of an ANSI-compliant LED standard, using UL's outline of investigation (OOI) document as a starting point. The purpose of UL's activities is that LEDs used in any type of lighting products achieve the same levels of acceptance and consumer confidence as traditional lighting technologies, especially with regard to the safety issues of risk of fire, risk of shock, and biological hazards.

In March 2006, DOE hosted a workshop to convene all key standards-setting organizations. The group reviewed LED standards and test development needs, assessed the development process and impacting time lines, and chose the Energy Star time line — which can be downloaded from at http://www.netl.doe.gov/ssl/PDFs/ENERGY%20STAR%20SSL%20Final%209.12.07.pdf — as its development goal because the energy-efficiency criteria are needed for the program to set benchmarks for LED lighting systems. The ANSI and IESNA working groups anticipate the release of new standards in early 2008, in time for use in the Energy Star criteria, effective Sept. 30, 2008.

An important point stressed by everyone involved in the LED research and testing is that while these standards will accommodate the characteristics of LED technology, they must still be valid for comparison with other technology.

“You've got to be careful in developing standards,” Narendran says. “Bad standards can hurt the industry. We need standards that are technology independent. If I am a lighting specifier trying to specify a lighting system for an application, my options are LEDs, halogen, and CFLs. I can have multiple choices as far as technologies are concerned. So whenever I compare a lighting system for a given application, I should be able to compare and select one independent of what the technology is. So the standards we develop have to be able to be used across all technologies — otherwise it could confuse the industry.”

Standards must address inconsistencies within a certain technology as well. Arnold uses ballasts as an example of successful cross-pollination, so to speak. “If you need to switch out a fluorescent ballast, you don't need to concern yourself too much with who made it,” he says. “That's not yet true with LED drivers. But we're getting closer to that.”

There are several standards under development by different standards-making bodies. Currently in committee review are LM-79 for luminous flux; LM-80 for lifetime, or lumen depreciation; ANSI C82.XX for electrical safety; and a standard for SSL definitions that has the standards organizations working in cooperation.

“You have to get these things defined so people use the same vocabulary,” Brodrick says. “It brings a common currency and a foundation to the market. People can see that the product's passed this or it's tested to that standard, so it's consistent and can be compared to others. It brings order to the market to have all these standards done.”

Sidebar: Product Performance, Measurement, and Safety Standards

The American National Standards Institute (ANSI), Washington, D.C., www.ansi.org, oversees the creation, promulgation, and use of thousands of industry norms and guidelines, including the following key standards of relevance to solid-state lighting (SSL) products.

Sidebar: In the Cold

C78.377, “Specifications for the Chromacity of Solid State Lighting Products,” will specify the recommended chromacity ranges for white light LEDs with various correlated color temperatures (CCTs) and ensure communication of chromacities to consumers.

“LEDs in freezer cases do very well,” says Nadarajah Narendran, Ph.D., associate professor at the Rensselaer Polytechnic Institute (RPI), Troy, N.Y., the director of research at the Lighting Research Center (LRC) at RPI, and the organizer of the Alliance for Solid-State Illumination Systems and Technologies (ASSIST) program.

According to a study conducted by the LRC and published by the Society of Photo-Optical Instrumentation Engineers (SPIE), Bellingham, Wash., human subjects preferred a display case with LED lighting over one with fluorescent. The study revealed it would be possible to develop an LED-based lighting system for commercial refrigerators that is competitive with fluorescent lighting in terms of energy efficiency.

“LED [lighting] is becoming real interesting in one of our biggest markets, and that is the retail display lighting for freezer cases,” says Randy Breske, CLMC, CLEP, president of the International Association of Lighting Management Companies (NALMCO), Des Moines, Iowa, and VP of Stay-Lite Lighting, Pewaukee, Wis. “There are some new appliance manufacturers that are going to LEDs, and they're working very well. They've also got a nice retrofit product that you can actually upgrade existing older cases with the LED technology. That's something that we're working on with some of the grocery stores we work with now.”

This is good news for supermarkets, which spend nearly half their annual electric cost for refrigeration. In addition, studies have shown that lighting accounts for about 15% of the total energy consumed by commercial refrigerators — not to mention a sizable reduction in maintenance with LED system use. “Fluorescents have some lumen depreciation over life, so stores that are real proactive in their lighting management are actually relamping all those freezer doors every 30 to 36 months,” Breske says. “With the LEDs, you don't have that group relapse. There's a double benefit in that regard.”