For those who still believe LED bulbs to be the perfect incandescent replacement, be advised that leading lighting industry representatives don’t seem very convinced themselves. Here are some of their views ‘straight from the horse’s mouth’ in 2009 and 2012.
In May 2009, EDN’s Designing with LEDs seminar invited Francis Nguyen, senior product marketing manager at OSRAM, Willem Sillevis-Smitt, director of strategic marketing at Philips Lumileds, and Paul Scheidt, product manager at Cree Components, to discuss LED issues and the future of LEDs. Quoting part of the article LED lighting: panel debates quality versus cost here (emphases added).
Conference chair and EDN editor Margery Conner opened the discussion with a simple question—why can’t we just use banks of ordinary 5mm industrial white LEDs for lighting?
The answer turned out to be less than simple. For one thing, Scheidt said, a typical 5mm LED produces about 6 lumens. So a big commercial fixture would require thousands of LEDs, creating interesting problems for driver designers, and in the end consuming about the same power as a florescent fixture of the same output. “The big advantage for LEDs in lighting applications comes when you design the LEDs for that purpose,” Scheidt said. “Then you can achieve output efficiency, directability of the beam, and reliability that you can’t get any other way.”
Nguyen offered a cautionary tale. “I walk through a neighborhood that has lots of those solar-powered LED yard lights,” he said. “And I started noticing recently that more and more of them are burned out. Well, those fixtures actually do use standard 5mm white LEDs. What happens—you can see this if you take one apart—is that the blue light from the LED is intense enough, especially if you overdrive it, to degrade the epoxy encapsulant. The epoxy turns yellow, and then it starts to crack. Once you get a crack, the LED will fail. So this is not a good use of these LEDs.”
The next question was more challenging. How will LED lighting avoid the same sort of problems that compact florescent lamps (CFLs) experienced early in their market life? Sillevis-Smitt weighed in first, explaining that like LEDs, CFLs had started out as high-priced lamps designed for long life. But the lure of the incandescent-bulb replacement market, coupled with the entry of low-cost producers from Asia, quickly drove vendors to trade away the long life in exchange for lower cost. This led to a complex situation in which consumers often don’t understand the characteristics of the lamp they are buying. In fact in many applications, inexpensive CFLs aren’t really direct replacements for ordinary light bulbs at all. But not knowing this, consumers buy the wrong bulb for their purpose and get frustrated when it doesn’t work. That is exactly the situation regulators are now trying to prevent with new standards, Sillevis-Smitt said.
Nguyen added that first of all, designers needed to understand that LED lamps required conductive cooling, while incandescents are designed to survive with only radiative cooling. So LEDs physically could not be screw-in replacements for incandescent bulbs in all applications. And today, he said, the cost per Lumen of LEDs is very much too high to replace incandescents or even CFLs. “Fixtures must exploit the long life of properly-designed LED lighting to have a competitive position,” he said.
Scheidt agreed with both statements. “The EnergyStar standards will be critical to the development of this market,” he said. Also, he underlined that the strength of LED lighting is not its cost, or even its efficiency, but the features that it can offer beyond just illumination. “We have to break people away from the expectation of a screw-in light bulb for very low cost,” he said.
A series of questions asked about what the maximum light output of the lamps really was, and whether it was a function of lifetime. Sillevis-Smitt answered that getting the best lifetime out of the lamps meant reading the datasheet, where output-vs.-life data was available. “In general, higher junction temperature and higher current will mean lower life,” he warned. “There are several failure modes on LEDs, but which one is most likely depends on the architecture of the particular lamp.
Doesn’t sound at all as reassuring as when the same lighting technology is presented to the general public with statements like “lasts 25 years”.
The article author (rather astonishingly!) concluded:
“All in all, the panelists seemed confident in the future of LED lighting, but definitely not as a screw-in replacement for incandescent lamps.”
So, how does the lighting industry feel about it now, three years later, when much progress actually has been made in the LED development?
This is THE scientific conference on lighting sources (lamps) and is held every other year. This year, it was hosted by Rensselaer Polytechnical Institute (RPI) Lighting Research Center (LRC), and held at the RPI campus. In recent years, this conference has focused heavily on LED and OLED — so-called “SSL” — sources with only a minimal amount of discussion centered on fluorescent, HID, halogen, and other light sources.
The chief technology officers of several important lighting companies (Osram, Philips, GE, Cree, Panasonic, and Toshiba) all spoke on the first day, and, surprisingly, I observed none of the hyperbole or excessive optimism about LED and SSL we’ve experienced in recent years.
In an ensuing panel discussion, one question from the audience was, “Can you list and prioritize the technical challenges with SSL, and can you tell us what the key issue to solve is?”
Without hesitation or disagreement, they listed cost, thermal management, reliability, color quality, electronics, and form factor. Two of the CTOs agreed that the top two are extremely serious, and necessary to solve before SSL can make significant inroads into the residential lighting markets. They all agreed that SSL is only relevant at the present for decorative and specialty lighting, especially backlighting. They also agreed that they feel major improvements may come about (ranging in opinion of 1 to 10 years.)
[leaving out part where everyone showed interest in the Vu1 bulb, as irrelevant for this particular article]
With regard to other conference highlights, the consensus is that OLED technology is far too dim, unreliable and too expensive for products at this time; but some years in the future, there should be a crossover in “cost per lumen” which could help them emerge as residential products.
The LED “Blue Light Hazard” is no longer viewed as unsubstantiated fodder for emotional hysteria: all of the LED community is now acceding to the existence of problems with circadian rhythms, melatonin and other hormone production, macular degeneration, disrupted sleep cycles, and other issues, as a result of exposure to LEDs – the potential to be a major worldwide issue.
OK, this description was from the producer of a competing lamp technology, but I have no reason to believe it inaccurate, especially not in view of the earlier panel discussion.