Lighting/Illumination Forum

In preparation for the LED revolution, knowledge about SSL has increased at all levels, from lighting manufacturers to consumers

CONVENED AND MODERATED BY RICHARD COMERFORD

When the annual Lighting and Illumination Forum convened last month, all participants seemed very optimistic about the prospects for solid-state lighting in the the coming years. But forum members had different views on how the transition would occur.

Electronic Products: Welcome to the 2010 Lighting and Illumination Forum. We’ve gathered a group of experts, some who have been here before, some who are new to the group. I’d like to do is start by asking each of you to introduce yourself and give a brief overview of some of the ideas you may have on the industry.

I’d like to ask each of you to look at some of the key developments in the past year that you think are important to driving LED lighting and how you think they’re influencing the industry, for good or bad.

Werner Mashig (LED Lighting Applications Engineer, Arrow Electronics) : I’ve been with Arrow Electronics for four years now. I’ve been in the solid-state lighting industry for about nine years and I’m an LED lighting applications engineer, a centralized resource for the company. I’m based out of Denver, CO, and our ultimate goal as a company is to accelerate customers’ time to market, transitioning them to solid-state lighting.

What we’ve seen in the past, especially early on, is a lot of customers transitioning from the incandescent bulb. This is a in a sense a no-brainer, especially because there’s going to be bans on the technology. You’re starting to see that over in California, and it’s probably going to adapt further across the country shortly.

But the key thing that we’ve seen recently is that customers are starting to request to start playing with LEDs and solid-state lighting solutions to actually replace more efficient incumbent light sources, such as the CFL and linear fluorescent. And so it’s a very exciting time for LEDs. I think a lot of that has to do with the LED advancements from the manufacturers on this call and others as well

And also I think that customers are really starting to learn a lot more about the technology and actually how to use it in their application. You have to remember this isn’t a type of technology that a lot of customers are used to, so I think they’re doing a good job overall. Everybody who’s providing these kinds of solid-state lighting solutions is educating the people to adopt more quickly.

Paul Scheidt (Product Marketing Manager for LED Components, Cree) : I’ve been with Cree for five years, so that means I’ve been with Cree before people thought LED street lighting was even possible, so it’s been an interesting road. Cree is now a leading LED supplier into lighting applications.

There’s a couple big-picture items that I’ll talk about first. First off is just the amazing amount of scale that’s been going on in the LED lighting industry as far as deployment, interest, and number of companies getting involved. I think that was readily apparent through Light Fair earlier in May in Las Vegas — how many companies are evaluating LEDs, coming out with LED-based lighting products.

Another big-picture trend that we’ve seen over the past year has been an explosion in the what I call the LED light bulb space, or the replacement lamp space. These certainly have become a very key and large market in the LED lighting industry, when I’d say there were questions just a year ago as to whether they were actually going to be technically viable. And now it’s one of the biggest areas of development you see these days.

So those are two of the big-picture items. There’s been a number of innovations that have gone on in the industry in terms of brightness improvement, in terms of color consistency, and in terms of standardization.

George Henry (Chief Engineer for R&D, Microsemi) : I have been with Microsemi for 16 years in various positions in Engineering, have been involved with lighting for 15 years, and started working with LEDs about 3 years ago. My primary role is to develop new technologies and new ways to drive LEDs. I’ve focused in backlighting over the years, and now am making a transition to solid-state light.

In terms of drivers, we’ve seen the industry rush to market with conventional power supply solutions. There has not been much concern for long-term reliability or special features. It’s been get to market with a very low-cost product.

We introduced our first LED-lighting driver this year and we’ve had very good response. It’s a very high-end product, but not that much more expensive. We’ve had major companies come to us to get custom drivers of this quality. So I think quality is finally becoming an important thing, and we believe here at Microsemi that lifetime of the driver must equal the LED, and it’s getting very, very good. Our long-term focus is on quality , control, and communications.

Electronic Products: Very important issues, yes, both the reliability and, of course, cost is also a large issue. And I think we’ll discuss that more as we go on. But let me ask Marc now to give us his perspective from the last year.

Marc Dyble (Product Marketing Manager for solid-state lighting, OSRAM Opto Semiconductors) : I’ve been with OSRAM for five years in both application engineering and product marketing, to facilitate the adoption of advanced LED lighting technologies over standard incumbent sources.

Whether large or small, traditional lighting companies that have not evaluated LED technology in the past are now taking a closer look. This is reflected in the increasing number of lighting luminaires using LEDs, as seen at industry trade shows and fixture manufacturers’ catalogs.

With respect to the LED, the general trend has been decreasing cost and package size with an increase in efficacy. LED adoption has risen based on several factors, including stimulus monies and The American Recovery and Reinvestment Act (ARRA), both putting focus on sustainable and energy-efficient lighting sources.

Another trend is new methods of creating white light, on both the fixture and component levels, leading to higher efficiency, color rendering index, and color consistency. Coupled with positive adoption of standards and regulations, the quality of available LED products is higher than ever.

Electronic Products: Yes, all of those are important, Marc. And I know CRI is one issue that’s going to be more and more significant.

Peter DiMaso (Strategic Marketing Manager, Lighting Power Products, Texas Instruments) : I’ve been involved with LED lighting for about two years now and I’m very excited about this. TI’s Lighting Power Products’ sole purpose is to provide the LED light market with solutions that don’t exist today. We are focused on both the driving of the LEDs and the intelligence behind them.

I agree with everything that’s been said so far, and those have been the key things that have been elevating the whole LED lighting industry. The most important, in my opinion, is what the people at Cree and OSRAM have been doing to get performance out of the LEDs and reduce costs.

That enables companies like Texas Instruments, at the component level, to create products that are going to take advantage of these improvements in LEDs, as well as what Microsemi is doing at the full driver level.

Another key thing I’ve seen [in the last year] is the general education of the solid-state lighting market about applications. I think a lot more consumers now are aware of the benefits of LED. And more importantly, a lot of the traditional lighting fixture people who originally did not have much electronics experience are getting to the point where they need a lot less hand-holding, if I can use that term, to transition from traditional light sources, whether it be gas discharge or incandescent, to LED lighting.

So these are things that I’ve seen over the last year. You can go into any city now in the United States or Europe and there will be at least one area lighting application, outdoor lighting application, that is using LED somewhere in the city. So it’s happening everywhere.

Electronic Products: In the New York area where I live, the Bayonne [NJ] Bridge is now lit by LEDs in a patriotic red, white, and blue theme. But all over the nation — the United States — and Europe, and Asia, we are seeing LEDs come to the fore.

Peter DiMaso: I’d just like to comment on the Bayonne bridge. I actually saw it. I got into Newark last night. It’s quite amazing, actually. It was really impressive.

Electronic Products: I believe LEDtronics supplied the bulbs, but I’m not sure whose LEDs they were using. It is amazing. When I see such LED lighting, I’m always amazed at how really rich it can be, how bright it can be, and how much it changes the nature of structures, making them stand out much better. And what it does for just general street lighting now, it’s so far away from the early days of harsh blue/white lights. It’s truly amazing.

But to start off, I’d like to ask all of you to discuss the LED revolution, what’s happening with market adoption, and what’s happening with user knowledge in this area. And talk about the adoption of solid-state light, as opposed to things like most recent competition from compact fluorescents and the cost aspects of this moving forward. What is a replacement LED bulb going to cost?

Marc Dyble: To the point of market adoption, I’ve seen numerous instances where LEDs are filling sockets traditional reserved for legacy light sources, including a greater number of highly-visible installations.

A recent LED downlight installation at the University of Notre Dame showed not only a decrease in energy usage, but student feedback indicated they looked and felt better under the light source and preferred to be in this environment lit with LEDs. Just like lighting the façade of a bridge, college campuses are no exception where a lighting designer’s image of mood and space can be achieved in a sustainable manner while still reducing energy costs.

Encouragingly, the knowledge of LED subject matter of the end user has increased over the past year, from not only educational programs offered by LED manufacturers, but through industry trade groups and organizations. At OSRAM Opto Semiconductors, we are an educator and enabler of the technology because it’s not a light source people are commonly used to. We’ve created a separate website focused only on LEDs for the use in general lighting applications to close the gap between the lighting and semiconductor industries.

Electronic Products: George mentioned the idea of reliability as being key to what he is doing. That’s much different than what we’re used to with incandescents. But user knowledge has been increasing ahead pretty rapidly from where we were, say, a year or even two years ago, when people were saying there had to be a lot more education of users about solid-state lighting.

Peter DiMaso: I agree with that, and we have to thank the Department of Energy for actually taking on the solid-state lighting marketing at a what I’d say is an international level, not just a U.S. level. They’re really setting the trend and doing a lot of good work with the CALiPER Program, with testing claims that people are making with their LED lighting products, so it’s holding the manufacturers of LED lighting accountable to proper advertising. The Lighting Facts Label is another great program that they have.

And now, at the most recent workshop from the Department of Energy, they talked about a grass-roots education program that’s going to further help the general public understand the values of LED lighting. This is really what’s helping the revolution. I think it took a few key people in the industry to really latch onto it and help drive the revolution.

Electronic Products: I agree, and I have to acknowledge the incredible work being done by Jim Brodrick of DOE in making everybody aware of this and helping organize these events. He’s done a wonderful job in this area. The DOE programs are moving things ahead very rapidly.

Paul, what’s your view on this?

Paul Scheidt: As far as LED adoption, we’ve seen a lot of that in the outdoor lighting space. The communities that have already adopted LED lighting want to spread the word among that community of users, among other municipalities. Cree has fostered that with the LED City Program, which you’ve probably seen is now wrapped up into the DOE Municipalities Program.

Municipalities that have integrated LED outdoor lighting in their community are excited about the change that’s brought in terms of how people feel about the spaces that are under LED lighting as well as the energy savings and maintenance savings that have come true based on the installations.

So there’s definitely been a big interest in this past year to accelerate efforts to get the word out among municipality groups to adopt LED lighting. I’d also say another thing about the adoption of LED lighting is that a precursor to adoption is availability of product. And we’ve seen major advances, especially to consumers in this space from a year ago in terms of products that are available at a Home Depot or a Wal-Mart, those types of avenues that people traditionally buy their lighting products. There’s a lot more LED options that are on the shelf this year than there were last year.

The initial cost is still a little bit of sticker shock looking at the first round of products. But as we all know, the prices continue to fall in the LED space, and I think it’s now going to be a matter of time before people start adopting more and more LED lighting products into their homes.

And I think the cycle will be pretty much the same that we’ve seen with the municipalities: once people try it, they like it and they’re going to tell their friends; they’re going to have people come over to their house, see it, and get a good impressions of LED lighting. And that will help start that snowball of adoption in residential environments.

Electronic Products: Absolutely. In fact, one of the things I’ve been impressed with, with some of the bulbs I’ve seen, is not so much what an individual bulb does, but the consistency from bulb to bulb in the quality being produced, to the point now where you can buy two, three, four bulbs and the color quality of the light, the brightness of the bulbs, is consistent from bulb to bulb, which is a major step ahead.

And it’s wonderful, of course, to be able to turn a switch and have the light come on instantaneously, as opposed to what you see with many CFLs.

But with regard to the improvement in education and knowledge of users, one thing I’ve heard from the DOE is it seems there are levels that are out there. There are some people who are really coming up to speed very rapidly and there are some who are just getting introduced to this. So in the user community, there seems to be a pretty broad range of levels of knowledge about these products. Is that what other people have seen as well?

Werner Mashig: I think it’s absolutely critical that, when you walk around all these LED shows and lighting shows across the country, you see all the hype. And people might kind of understand what’s going on, that it’s the cool thing to do and that it might be the future.

But I think it’s actually critical definitely from all of our standpoints that we actually get products in front of the customers to play with because I think that they will learn quicker by actually testing the products and understanding on how LED systems work — that they’re constant current sources; that they conduct heat out the back; that they’re quite different from traditional light sources.

So I think that within the LED revolution, in order for it to absolutely explode, people have to learn by doing, as well as doing their research and continuing to get the cost down. And LED and driver companies continue to advance their technology on the rapid pace that they continue to do so.

Electronic Products: I agree. Anyone else care to comment on this area of market adoption? I’m, again, Paul alluded to sticker shock, and I think that probably does have some effect, paying $20 for what one was used to paying maybe $1.50 for in the past must push back some consumers in this area. On the other hand, the municipalities seem to get it, seem to see the idea that what they save in terms of maintenance, what they save in terms of power, really makes a difference.

In getting to the consumer marketplace, what do you think the key elements are? Anyone care to comment on that?

Marc Dyble: Color uniformity and consistency between lamp to lamp is very crucial. When the compact fluorescent lamp was first introduced, they suffered from poor color, not to mention noisy electronics noticeable flicker. Even with technology improvements and marketing campaigns over the years, CFLs today are still perceived as an inferior light source. ,Getting past that point requires putting product in the hands of the consumer, to determine the quality for themselves,

I think it’s being done very successfully, especially in the residential space where LED adoption is on the rise due to the positive technology attributes.

Electronic Products: Speaking about the attributes, one of the things that may come as a surprise to some people is how the bulbs will work with your traditional triac dimmers. I’ve seen some solutions that seem to be quite good and, in other cases, maybe there’s not quite the kind of dimming you’d like to see.

Can we address that issue of dimming and what we see out there in dimming for solid-state lights?

Peter DiMaso: I’ve got to say that being able, at least for a replacement bulb application, to get that dimming right is the holy grail. And it really is a challenge.

To just briefly to explain, traditional triac dimmers require a lot more current to be triggered, to be held, and to work with the EMI filter. With LED lighting, the total amount of current that going through these triac dimmers is a lot less and we have to do tricks to keep the triac itself happy. And whatever you do, there’s going to be an impact of some sort on performance, whether it’s going to be an impact on the driver’s efficiency, or it’s power factor, or the total harmonic distortion that you get out of it.

All these are going to be affected by dimming, so it really has been tough for all the industry.; I’m not just speaking for TI; I think anybody who’s been working with triac dimming has had some issues. As you said, you can pick up any bulb, put it on a triac dimmer, and there are very few that don’t flicker. There’s more that do flicker than don’t out on the market today, at least from our research.

But what is dimming? Dimming’s a pretty big definition, really. “Does your light bulb dim?” “Sure it dims.” But what does that really mean? It dims without flickering? So what NEMA has done is come up with a white paper. I’m sure you’re all familiar with it (LSD 49, I believe it’s called) and it’s actually talking about best practices. But they’re looking to make that a standard, where they’re going to take dimming and they’re going to make sure you specify everything, like “what is your maximum light when your position is fully on?” And “What is your minimum light at maximum dimming position?” And “What is the profile of travel throughout the dimmer?”

And flicker is going to be one of the specs, but it’s only a small part of what dimming is. Right now, people tend to consider an LED light bulb that dims is one that doesn’t flicker, but it’s a lot more complicated than that. A lot of effort’s been put into tackling those issues and that’s what a lot of the new products that are coming out are going to be addressing.

Electronic Products: Interesting. Would anyone else comment on that? George, do you have a view on that, on dimming?

George Henry: I’m starting to develop a view on that. I’ve looked at a few dimmers that are on the market. I agree 100% with Peter. The average consumer might not know all that technology and use all those terms, or even look at all the features on the product packaging, but he is sure going to notice poor dimming performance.

I’m one of these guys who has CFLs in my closet and every morning when I turn those babies on, they flash and then they glow pink for five minutes before I can finally see to select my shirt for the day.

Electronic Products: Not what I expect from a light bulb.

George Henry: So Peter, one more question for you. Do you believe that we actually have to use these triac dimmers forever, or at some point are we going to be able to replace them with dimmers intended more for LEDs?

And another part of the question is, do you believe in the bulb replacement market, that we’re going to need to be compatible with the old Edison socket in the home forever?

Peter DiMaso: How much time do we have? [chuckle] Those are great questions, George. So do we have to be compatible? I think, undeniably, yes because there are, depending who you listen to, anywhere between 150 million to 1 billion installed triac dimmers. You know, mainly in North America and in Europe. But they do exist, so we’re going to have to be compatible with that.

But in the future, I do believe that there will be a dimmer that’s going to be specific for LED lighting. It just has to happen because, right now, whatever you’re doing, it’s a compromise. Either you’re throwing away power, or you’re affecting the power quality further down the line. If think that if you match the dimmer with the end light source, you’ll have a much better product — more efficient and cleaner.

So I do see that happening, but the reason we haven’t seen it yet is because the Lutrons and the Levitons of the world, they feel that, “Well, what’s the use of introducing a new dimmer if we don’t even have enough traction with the LED light source at this time?” And are you just going to be increasing the barrier to adoption? I tend to agree with that philosophy, so if we introduce something now, it’ll probably be too soon for the market. But in the future it will come.

And your question about the Edison socket. In my opinion, light bulb replacement is just the first step. This is part of the education process, of what we can do with LED lighting. And this will add to the revolution because, once people see that, “Hey, all of the sudden I can replace my light bulb and save about, oh, close to ten times the energy and it’ll last ten times as long.” And then comes “Wait a second, you know, maybe we can do some more stuff with this.”

And that’s when people are going to get a lot more interested in what you can do with it and different form factors are going to come out. Twenty years from now, I expect new homes and offices that are built will have luminaires without Edison sockets because people will be replacing them with LED lighting.

Electronic Products: I like to point out that, when Edison was first bringing light bulbs into the home, what he did was say, “Okay, you can install this through the existing gas system that you have, and you can run the wires through the gas pipes. And you can put the Edison socket on top of where you put the mantel for the gaslights.” But even more ironic is the fact that his idea was to run dc. If that had succeeded, where would we be today with solid-state lighting?

Peter DiMaso: Yeah.

Marc Dyble: We have to certainly thank the professional and trade organizations for past standardization of sockets and interconnects and for what the future might hold. ANSI has several working groups defining these new generation of socket interconnect, determining how light subassemblies will fit together for LED luminaires and what these interfaces will look like.

Safe, low-voltage and scalable dc-powered systems, such as what the EMerge Alliance is supporting, could be a new means of distributing power in commercial buildings. OSRAM is working with several companies to deploy direct-current architectures driven off battery or renewable power sources for LED lighting.

Paul Scheidt: I want to comment on the LED light bulb form factor. I can’t take credit for this idea (it was Mark McClear’s), but think about another technology. If you wanted to get to play music out of an iPod on your car stereo back before everything had iPod connectors, you had to use that cassette tape adapter. Remember that? And so that’s when you had to jam one new technology into an older technology in order to make it work in a system. But then very quickly it gets replaced. Because, like with the iPod connector, you can add more functionality.

So I think this piggy-backs on what Peter was saying, that the application of new controls and new functions to lighting systems will draw people away from that existing light bulb form factor into other form factors as they realize what more could be done with the technology.

Electronic Products: I wonder how much of that is going to depend on new construction, versus retrofitting new forms into existing structures. That’s part of the tradeoff. Look where we are in the home building market today; look at what’s happening in that part of the market. There’s an awful lot of housing stock out there right now, so that building new homes with new systems in them may not happen as quickly as we’d all like it to see it happen.

But I do have to agree that the retrofitting has always been a transition stage. Nobody today goes around installing gas lines so that they can run wires into a fixture. I think that the LEDs will create a whole new way of creating lighting, including new types of sockets, because the issues that you deal with with LED lighting are different, much different, than you deal with incandescent. The way the heat is radiated is particularly critical.

One of the things that’s kind of interesting to me is, considering how much less energy the LEDs use than traditional incandescents, you wouldn’t think that the issue of heating would be as important as it is. And yet the heat points are so concentrated that it becomes a very hot, intense source of heat. “Does it have to be that way?” is the question. How do we deal with heat in LEDs?

Marc Dyble: The solution is highly dependant on the system’s thermal design. Once the system is powered on, localized heating will occur at the center of the LED package. Thermal techniques, many adopted from the electronics industry, can be applied to spread this heat away from the LED to a more optimal location, where it can be conducted or radiated away. Until 100% conversion efficiency is achieved (and no heat is generated), then the thermal design is still a critical factor..

Werner Mashig: I want to comment on the previous topic and the new one. And I think it goes hand-in-hand. You know, this is exactly why retrofitting LEDs in something like an A19 type form factor is really not ideal at the moment. You can’t find a 100-W LED replacement because the actual amount of footprint you have to heatsink the LEDs in a bulb is just not there. There’s just not enough mass available in that form factor to basically dissipate the heat out of the LEDs.

Short-term, I think we’re going to have to adapt to what’s out there: the triac dimmers, the sockets, et cetera. But long-term, it definitely makes a lot more sense to design luminaires and systems that are meant more for LEDs.

And this goes for the optics as well. What you see out there in the traditional world is a lot of reflector-type systems, it is a very ideal solution for today’s optics.

However, if you put a lens over an incandescent bulb, it’s just going to basically melt, especially if it’s plastic. So now, when you bring in the LED, you can actually think of different ways on how to get the light exactly where you need it, instead of just automatically going to the traditional reflector type of system.

Peter DiMaso: True, we are limited. We cannot get a 100-W equivalent out of an A19 today. But to diverge a bit, I’m living the dream. My house is full of LED light bulbs right now in the A19 form factor. (Please do not tell my wife how much I spent on them.)

My point is that these are only 390-lm bulbs, so they’re nowhere close to being even a 60-W equivalent. However, I can’t tell the difference from a 60-W equivalent and this 390-lumen LED bulb; I am getting enough light for what I need. So to me, that’s kind of a shift, a paradigm shift, really. Do we need to replace the 60-W and the 100-W with a functional equivalent, or are we getting enough light with this new technology?

Electronic Products: That’s an interesting question.

Werner Mashig: That’s a really good point, Peter, because what we’re actually seeing is a lot of people kind of look at the light source’s total efficacy and lumens, but when they actually put it in a luminaire, a lot of these incumbent light sources actually do very poorly in fixtures, and there’s a poor coefficient of utilization. So that fact that LEDs are more directional, you can get the same amount of light on the ground with a much lower power, or you can actually go with the same power and get a lot more light.

And like you said, I think a lot of it is all about how you perceive the light. In a lot of cases, you can actually get less LED lumens on the floor and it actually looks the same or better just because of the color quality over maybe some of the more efficient incumbent sources, such as like a fluorescent.

Electronic Products: Sure, LED light sources are not omni-directional like incandescents, but there are so very few real applications that need it. The ceiling troffers here in my office are pushing the light down with reflectors onto my desk, rather than shining it out 360 in every direction. An important aspect of solid-state lighting is that you can actually have directional control. It’s built into the nature of the lighting. And usually you want the light where you want the light, not all over the place.

Marc Dyble: This can be described by the term “application efficiency”, or how well your light source puts light where you want it. In some cases you can use less source lumens (light) by efficiently directing it where you need it, and not throwing light away by scattering it in all directions.

Electronic Products: Right. It’s interesting.

To shift a little bit here, let’s talk about testing of LEDs and the testing of luminaires and how these things actually differ. What do we need to do today to check out and validate LEDs and luminaires?

Marc Dyble: The IESNA has a testing procedure committee that’s developed two standards for this purpose. One is LM-79, which is the electrical and photometric measurements for solid-state light products, a system-level testing standards. The second, LM-80, is a measurement of lumen maintenance of the LED light source, which is a component-level testing standard.

Electronic Products: Are the testing standards pretty well down the line; is there good agreement on how you should be testing the performance of these devices?

Paul Scheidt: Yes, there’s a consensus agreement within the LED component industry that everybody latched onto, LM-80 being the standard. I don’t see that being an issue. Everybody understands that those are the standards by which the LEDs are tested.

I think there’s a higher-level question that’s going on further down in the industry: How long exactly are LED lighting products using those LED components going to last? It’s a question that comes up quite a lot and that’s a much harder question for everybody to answer. It really boils down to two things. First off, the LEDs that have been made in the last two or three years, by and large if they’re coming from a reputable supplier, are fairly reliable.

And so what happens when we do this LM-80 testing is we’ll test for two or three years on a component, and the LEDs will only degrade a few percent. We define end-of-life as being a 30% degradation, and even if we test them under high temperatures, under high current for a long time, it’s still not getting anywhere close to that 30% degradation.

So at a component level on the lumen maintenance question, it’s somewhat difficult for the industry to answer in a believable, consistent way because we just have not seen that much degradation. We don’t have more data. We don’t have perfect vision into what the future’s going to hold for these parts. But at least the data to date has all been proven it to be very reliable.

And so that’s where I think this question even come up is to say, “Okay, if the LEDs are going fine, then what about everything else in the system?’ And that’s where it gets back to a little more focus being placed on luminaire testing saying, “Okay, if we give the LED the free pass, then is everything else in the system going to last for the rated lifetime on the box?”

And I think everything that has been developed has been developed for good reason, and put in place for good reason. I think, because LED quality had gotten so much better quicker than anybody ever imagined, the question has now changed, like I said, from focusing purely on the LEDs to the complete system.

Peter DiMaso: I’ve got to say Paul hit it right on the head. The luminaire testing really has to be where we move forward. And that’s where a lot more is placed on the driver itself, so that scenario is where I guess George and myself and Werner are even more concerned, because that effects the products that we’re developing.

To answer your question about, “Are the standards pretty well set?” Actually, no, they’re not. I think we found out recently that there’s a loophole in the way that you can do some MR16 voltage testing, in that you can actually apply a dc voltage to test the efficacy and you’ll get much better numbers than you would actually using an ac voltage. So I still think there’s a lot of work to be done in the standards to get them up to what they need to be.

Electronic Products: Are there specifications for the other components that go into the luminaire, the other electronic components? Are they comparable to the ones for LEDs? I mean, if you would specify an operating temperature range, is it the same operating temperature range in all cases for the electronics and for the LEDs?

Werner Mashig: Regarding the general feeling around the electrical part of it is, well, everyone says, “Electrolytic capacitors are the weak link in the system.” And that might be true from a specmanship standpoint, but again, to everyone’s point, it’s all in the design.

Basically, with a really good luminaire design — by spacing out those electrolytics, keeping the heat away from them, maybe over-designing them, in a sense — you could get as much lifetime as you can with the LEDs. Like we said before, “One bad apple spoils the bunch.” If you have a bad thermal design, the whole luminaire will fail quickly.

If you don’t design the LEDs properly, the drivers properly, you know, to George’s point before, and if cost is everything, it’s going to be very difficult to really set yourself apart from the rest of the industry, as far as a good luminaire is concerned, and [provide] something that will last a really long time.

Frankly, I’m a little worried.You know, this happened with compact fluorescents. They had a couple hiccups in the beginning. LEDs take more work to basically design a system, especially something that the lighting designers really aren’t familiar with. Hopefully, we won’t have any hiccups moving forward with LED lighting design in the industry.

Electronic Products: So we’ll need strong standards and somebody like Underwriters Laboratories who can come along and say, “Okay, this does meet the specifications that we’re talking about.” And have that broadly enforced, with a seal of approval type of thing on all these products.

Peter DiMaso: Yeah, I totally agree. And as Werner said, you can find a driver that will not last 10,000 hours if it’s not designed properly, or you can get a driver that will be 50,000, maybe 100,000 hours if you do the proper work to ensure the proper derating of the capacitors and any other component that might be suspect.

Marc Dyble: From a thermal standpoint of the driver, little emphasis is placed on the driver componentry in comparison to the LEDs. I’ve seen systems in which the LED board temperature is reasonable but the driver components could be 20 to 30 higher than the rest of the system, which could be easily overlooked in some designs. This is complicated even further when the driver components are located in a harsh environment, such as inside the screwbase cap on the bottom of retrofit lamp, where temperatures reach high levels.

Electronic Products: Oh, interesting. Kind of goes back to our earlier discussion about the knowledge of the user base. Obviously, this is a key area where designers have to come up to speed and make sure that they understand the requirements for the luminaire design to make the electronics operate as efficiently as the LEDs do, as reliable as the LEDs do.

We mentioned cost again. Two questions I have are (a) what are the cost factors today that keep the price of solid-state light to be significantly higher than other traditional lighting? And(b) how do we bring those costs down? What affects the rate at which those costs drop?

Paul Scheidt: The most obvious answer to this, of course, would be the LED component and focus on that. And certainly things are being done on this side of the fence to get those costs down, whether it’s traditional semiconductor answers like going to larger wafer sizes or increasing capacity. All those traditional types of things are routes to lower costs for LED products.

But in the past year especially, I’ve seen a lot of creativity being done on the system level in ways to address system costs. And things like moving to higher-voltage LED arrays, and with higher-voltage LED arrays, getting closer to the line voltage, enabling more efficient, smaller, cheaper power supplies being one option in terms of system design.

Another one is as LEDs get brighter, you can use fewer LEDs. And with fewer LEDs, then you need fewer optics, if your system uses one optic per LED. So you may not be reducing your LED costs or LED spend, per se, but your total system cost goes down because you’re using less optics. So those are routes that I’ve seen being implemented fairly effectively in system design in order to help get the cost down.

Peter DiMaso: I agree with what Paul is saying. And the LED manufacturers have been keeping up their end of the bargain and increasing the amount of light out per dollar at a rate that’s following Moore’s Law and actually probably accelerated, versus what we’ve been seeing.

And what I have found is that the actual cost of all the driver components — and I’m not just talking about the controller IC that TI makes or anybody else makes, but the total driver cost — over the last year (although it’s been decreasing as well) has become a larger portion of the overall luminaire cost because the LED lights have been coming down in their cost.

And reason for that? There could be many, right? But an LED driver is very much like a power supply. And power supplies — surge and power versions — have been around for a long time, and the cost has been beat out of them already. So there’s still room for improvement [in LED drivers] and for looking at new typologies that could potentially save money.

But at the end of the day, if you reduce the cost of your driver, you will in some form compromise the performance of the drive current to the LEDs. So some of the things to consider is how much ripple current is acceptable, from an LED perspective and from a visual perspective.

Isolation, right? Where is the tradeoff? Is it better to go with a non-isolated driver and then do safety isolation at the mechanical level, or vice versa? That’s a cost tradeoff as well. And those are some of the things that have to be considered when creating the drivers. You can simply make a capacitive voltage divider off of the ac line, but the light quality’s going to be horrible if you do that.

Still, that’ll be your most cost effective way of doing that, and I would hate to see the industry go in that direction because we have a superior light source, and it would be a shame to throw it away with a poor driver.

Werner Mashig: This is a great topic. I’m excited to talk about this one because I agree with Paul and Peter that the majority of the cost seems to be in the LED components and the power supply or the driver. But one thing that I’d want to point out, too, is, you know, off-the-shelf power supplies, there’s many advantages and disadvantages to that. Just like the triac dimmers, this is something that the industry’s used to with maybe fluorescent installations where they’re basically hooking up a ballast to the fluorescent.

Similarly, you know, customers are comfortable with this, you know, model of basically, you know, wiring and LED constant-current power supply to and LED array. And that’s kind of their strategy on creating their luminaire design. However, this is a great approach if you need quick time to market and, you know, you don’t want to deal with too much UL approvals.

However, the long-term solution and maybe this is where we’re going — who knows — is I think is pretty unpredictable. But I’m just saying that, you know, if you design your own power supply, there could be various paybacks over time if you’re willing to invest in that UL recognition.

Electronic Products: Marc, Paul had said that he’d seen a vast improvement in the LEDs and the price points with the LEDs and the way in which they’re being used. Do you see that’s putting pressure on the driver, as Peter says?

Marc Dyble: I certainly agree. LED component manufacturers are generally reducing their price points through cost-down measures such as new LED packaging materials, larger wafer sizes, and managing yield. Today, for example, the LED content for a luminaire might be 60 to 70% of the bill-of-materials cost. Tomorrow I anticipate it being much less, possibly allowing for more complex controls or optional features.

George Henry: We are an IC company; we do modules, but we do them primarily to learn more about what we need to do in ICs. The reason us guys are in business is because we can create more value in a small chip than a guy with a discrete circuit or capacitive divider can.

I have actually seen guys trying to do capacitive dividers and it kind of does work. But as Peter says, we would hate to see the industry go that way. I think there’s opportunities for new topologies that actually do give better performance than the existing power supply structure. And they’re made possible in part because of the industry’s willingness to accept the tradeoffs between mechanical isolation and electrical isolation, for one example.

Electrical isolation is not cheap in cost, efficiency, and componentry, so that’s definitely an area we can look at. Also, integrating the control with the power conversion we think is critical for cost because in an IC we can do that very economically, to the point where the user doesn’t even notice the price increase. And so he gets the value of an IC because now he as all the features that LEDs promise without paying a significant premium for them.

Electronic Products: We’re getting close to our time limit, so I’d like each of you provide an idea of what you think the future holds in store for the next year, and even further out. We’ve been talking LEDs primarily, and some people are talking OLEDs also.

Werner Mashig: One of the things that I’m pretty hopeful for is to kind of goes back to the standardization. I think this is going to enable people to adopt much quicker and more on the same page. But one of the things that I’m really hopeful for is to straighten out this whole CRI perspective out in the industry.

You know, CRI is nothing but measuring your light source against the incandescent, and to me, the incandescent isn’t a perfect light source. I think the more perfect light source would be something more like the sun, which is natural light. So if anything, I’m hopeful that the LED in the future will be measured against sunlight instead of the incandescent bulb, because there are actually applications where LEDs actually look better than the incandescent.

Also CRI is based on 14 color points. If the world basically consisted of 14 colors, that would be the perfect test, right? But unfortunately, or fortunately, I should say, we have millions of colors in this world. And I think the sun does a better job of bringing out millions of colors than the 14 measured with the incandescent bulb. So this is one of the things that I’m really hopeful for in the future, to enable the revolution.

Paul Scheidt: One of the things we’re excited about is the concept of application optimization of LEDs. We looked at the traditional lighting in the industry and there’s thousands upon thousands of different bulb types. The lighting industry as a whole is very used to specialized light sources for specific tasks.

We’ve really taken that to heart and, now that we’ve got this broad portfolio of LED component, we’re figuring out how to take the different technologies at our disposal and mix them together in order to create LEDs that really are the best for specific lighting applications.

We’ve started down that path with products like the EasyWhite products we have right now that mix multiple LEDs together inside one package to deliver consistent color results. And those offer hundreds, if not thousands, of lumens of light in one package, so not only do you get consistent color, but it also comes in a form factor that you have a single point source of light. It also emulates a filament-type fixture like a halogen bulb, if you will.

So that’s really one of the main areas that we’re looking forward to in the next year. The other obvious one, of course, is performance improvements — making brighter, more efficient LEDs over the next year that enables fewer LEDs to deliver the same amount of light. That also addresses the cost aspect of getting things down. We’re also exploring different methods of putting LEDs together in order to offer ease of use and quicker time to market for our customers.

George Henry: Well, I think if this is going to be the year that the consumer market realizes the LED is here, then we need to give them some value. At Microsemi we are focusing on reliability and new typologies that will give better efficiency and better performance along with lighting control to easily optimize an LED for specific applications. This is where we’re headed.

Marc Dyble: New LED package designs are being introduced to meet the demands of our customers. Traditionally, LEDs were touted as a very flexible light source to work with, but a lot of times we see them being used in very traditional ways. By enabling new package styles to facilitate creative application, the look of a traditional light source might be something of the past.

New, emerging markets are also taking shape. One might be the horticultural and plant growth industries, where LEDs are very well suited by generating very specific wavelengths of light to stimulate plant growth, and saving energy by eliminating energy that isn’t needed to facilitate the growth of the plants. The light dosage can be prescribed based on plant type to optimize the growth of the plant and surrounding light conditions.

With that I also see organic LEDs (OLEDs) being used to add new freedoms when designing luminaires. Being a two-dimensional, diffuse light source, it’s very aligned well low-profile fixtures by not requiring additional plastic or metal to make up the luminaire casing.

OLEDs are expected to see the same increase in brightness and efficacy levels as LEDs enjoyed, over 100 lm/W, with the ability be transparent and flexible for application as window glazings or to wrap around curved surfaces. New applications of light are certainly on the horizon in the years to come.

Peter DiMaso: The LED lighting space is really a clean sheet. It can go in so many different directions, so I’ll try not to dream too much here and try to keep my feet on the ground.

But there’s so many ways that we can go. From an application perspective, the LEDs lend themselves much better to be remotely controlled, and used for other purposes, adding communications to luminaires and even to light bulbs for that matter.Can you imagine being able, through an app on your iPhone, to turn your lights on or off from anywhere in the world, because you forgot to do it before you left home. And if you have a light bulb that’s IP-enabled, can you imagine the ways you could go with that?There’s also occupancy sensing. And once you add the sensing, well, why not add a camera to that light fixture? You could increase security to a much higher level by adding the intelligence in those spaces, right?

From the very basic performance perspectives, I think we’re going to see a lot of improvements over the next year or so in the traditional dimming: the wall dimmers, whether they’re leading edge or trailing edge. And we’re going to see a lot more capabilities with zero-to-ten volt dimming. Again, adding communication with power line communication or some sort of wireless type of application.

Also coming in the future I see integration happening. Probably not in the next year or so, but we can see a lot more of the driver and the LED being integrated in a package for application optimization. I see a lot more application-optimized packages, with the driver and the LED co-packaged. In terms of retail shelf space, we’re going to see even more movement than we’ve seen in the last year.

I see a lot of synergy and good vibes in the market today. There’s a lot of collaboration going on. Here I am, online with a competitor, but I feel comfortable talking about where we’re going, because we all believe in the revolution of LEDs. We all want to make that happen, and I see that as an industry-wide phenomenon.

Electronic Products: It’s a huge market out there, and I think there’s plenty of room for everybody in it when you come right down to it. It’s hard not to start thinking in the future, of what could be, and what can happen with this technology.

It’s interesting that you mentioned the IP-addressable light bulb. I remember Vint Cerf back in the 1990s thinking along those lines. And 20 years later, we’re finally at the point where we can really make that happen.

It’s amazing what we can do with this solid-state technology — open up a vast number of new areas that had never been there before, never been possible before. Who knows what Edison would have made of this? I’m sure he would have had some interesting ideas about what you could do. ■