Dave Green had a favorite routine that he used in presentations to impress potential investors in his new company, Carmanah Technologies. The company's first products were navigation lights made from a combination of LEDs, solar cells, and batteries. Green would ask the biggest person in the audience to come forward and invite him to whack one with a hammer. Encapsulated in hard plastic, the lights were virtually unbreakable, which of course is an unusual thing for a light to be. The idea was to get everyone in the room talking, to focus their attention. Green had pulled off this stunt many times, and it had always worked like a charm. One particular night, however, in front of about fifty or sixty locals at the Union Club of British Columbia in Green's hometown of Victoria, things went awry.

“I got this big hulking guy to come forward,” Green recalled. “I gave him the hammer and I said, Here—bet you can't break this, it can withstand anything. He put the light on the floor and he whacked it, and this thing flew two or three feet into the air, and when it landed, he whacked it again, and it flew up again. I said, OK, that's enough. I took the light and I put my hand over it—it has to be dark before it'll light up—to show the audience. Well, the damn guy had broken it. The batteries had come right out, they weren't secure, I could feel them moving around inside. So I went behind the screen where I was doing my PowerPoint presentation and I got another one. Anyway, he hit this other light and he broke it, too. And by then everyone was roaring with laughter.”

To make matters worse, the Canadian Broadcasting Corporation was filming this debacle, as part of a series on entrepreneurs. The program was broadcast repeatedly. For years afterward, when Green met people, they would ask him what he did, and he would tell them that he made solar-powered lights for navigation. “Invariably they would reply, Oh, I watched a show on CBC about that. There was this guy, he broke his lights in front of potential investors. And I would go, Yeah, yeah, that was me. It haunted me for years.”

Nonetheless, Dave Green would have the last laugh. By 2005 Carmanah would be the world's leading supplier of solar cell-based LED products. In that year the company had sales of almost $40 million, more than a hundred employees, and an annual growth rate of around 70 percent, making it one of Canada's fastest growing firms.

Like Cree's Neal Hunter and Color Kinetics’ George Mueller, Dave Green is a serial entrepreneur. On graduating from the University of British Columbia with a PhD in oceanography, Green discovered that there were no jobs available in academe. So he and a couple of friends started what blossomed into a group of ventures, all of them initially with a marine flavor. Seastar Optics, for example, began by attempting to make a new type of oxygen sensor for seawater. This involved the use of fiber optics and a laser. It was hard to get the laser to shine down the fiber, so the team had to come up with a way to link the two. The coupler they developed would turn out to be of considerable commercial value for optical communications. But not in the short term.

In 1989, after fourteen years of working his butt off with very little to show for it, fed up with continually having to put his personal finances on the line to keep the underfunded firms afloat, Green decided that he needed a break. He sold his share in the businesses to his partners and went off to the University of the South Pacific, which is based in Fiji. He spent three happy years solving all sorts of problems for the islanders. One of the things that Green encountered down there that he had not previously been exposed to was renewable energy, especially solar panels. He used most of the money he got from selling out to buy himself a small sailboat, yachting being his passion. (“I love the complexity of yachting,” he told me, unprompted. “It completely engages me.”) After leaving Fiji, he and his wife and three daughters sailed around the South Pacific before heading back to Victoria, which they reached in 1993 after a nine-month voyage.

“By then I'd been out of the country for five years, I was forty-five, I had no job to come back to, and not much money left. When I came back, I found that my ex-partners were now wealthy. They'd sold the laser-fiber connector business to an American company. They had a lot of people working for them, they were doing great, whereas I was almost broke.” He started doing consulting work for the forestry industry, dubbing his consultancy Carmanah (pronounced car MAN ah, the name of a valley in a local old-growth, state-owned forest.) “But I wanted to get into making products again. So I went to [Canada's] National Research Council to get a small grant—fifteen thousand dollars—to look at the idea of a solar-powered anchor light for boats.”

It would be, as Green knew from his own experience, a great thing for sailboats to have self-contained lights that could run independent of batteries. Maritime law mandates that when your boat is moored or at anchor, you must show a white light with a range of two miles. In practice, however, yachties who leave their boats in a certain place for more than a day or two don't usually bother leaving a light on, because where's the power going to come from? Many was the morning out on the ocean that Green had woken up to find his boat battery drained and his navigation lights out as a result.

Solar panels and LEDs are beautifully matched. Indeed, it's as if they were made to go together, the one extracting electric energy from light, the other emitting light from electric energy. The combination of the two technologies would be key in everything that Carmanah did. The problem was that, back in 1994, there were no white LEDs available. Green thought that yellow ones might do the job, so he and an electronics guy he had hired to help him mocked up a prototype LED-based anchor light. “I put it on my boat, walked around the harbor, but I couldn't see the damn thing for more than about 150 yards. So that wasn't going to work.” But at least it was a start. Green and his engineer came up with the idea of a completely self-contained—“potted,” to use his word—solution consisting of solar panel, LEDs, plus a storage device known as a supercapacitor. Their idea was to invent a lamp that would make light forever. They proceeded to patent the design. “It's completely potted, so you can't change the batteries. Supercapacitors and solar panels and LEDs last forever, if you treat them right. So if you put one of these lights in, you'd never have to go back to service it. It would last you until you were dead.”

But nobody wanted to buy Carmanah's first batch of lights because they were not bright enough. “We tried almost giving them away, for making bicycle trails and things, but it just wasn't going anywhere.” Then came an unexpected consulting contract from the US Navy, to perform an environmental assessment on a nearby torpedo testing range. In order to gauge the performance of the torpedoes, the position of everything—submarine, tracking vessels, and target—had to be fixed. This in turn necessitated the use of buoys within the strait where the tests take place. “And they're huge, these buoys, they're hazards to navigation, they have no lighting, but they're never in the same place, so you can't mark them on a chart.” Following some nasty near-accidents, there had been complaints from the public about these buoys. “So one of the recommendations of the assessment was, You need to have lights on these things. The navy's response was, Well how are we supposed to do that? I thought, OK—I can solve that!

“We decided that we could make a red light, which was a lot easier than a white light, which didn't exist, because by now the high-brightness red LEDs were coming along nicely. So we produced a light, and that allowed us to expand the business. But red lights aren't terribly useful in the marine environment without a green.” Navigation lights indicate the orientation of a vessel, hence the direction in which it is traveling, using red for port and green for starboard. “When we started there was no green, but just after we finished making our red light, the green LEDs from Nichia became available. So now we could say we made navigation lights.”

Green showed the new lights to the Canadian Coast Guard. “They were delighted with them. They just couldn't believe these things, because before they'd been paying ten times as much for something ten times as big. But the combination of LEDs plus solar cells made everything much more efficient. Because you no longer had to have a great big buoy, which was needed to contain a great big battery pack, which then needs a great big chain to anchor it, and a great big ship to take it out to be moored. And when the batteries die, the great big ship has to go back out to the mooring place and lift the buoy out of the water to change the batteries.”

With Carmanah's new lights, you could use much smaller buoys that did not need a big ship to transport them, and that, once in place, did not need to be serviced. Because the current the LEDs draw is so small, they just use the “float”—enough current to equal the self-discharge of the battery. They don't drain the batteries, which basically stay charged all the time. You could also throw in a microcontroller—which, like solar cells, is a natural match for LED technology—to do various energy-saving tricks. Thus the annual maintenance costs for a navigation marker fall from as much as fifty thousand dollars per year to zero.

Navigation was a tough business to crack. “You put these lights out on a buoy in the middle of nowhere, they get bashed, they get soaked in saltwater. Some of the buoys get destroyed, and the bits end up on a beach somewhere.” But Carmanah's lights were up to the challenge. “They've developed this wonderful reputation for being indestructible. The lights are so self-contained, they have no on/off switch; I mean, there are no holes in these things. You use a remote control to turn them on, so there's nowhere to leak. We're now in our seventh year since we started making them, and they're still not coming back.”

LEDs are also much better than incandescents at pulsed operation. They turn on and off instantaneously, unlike the old navigation lights, which are filament bulbs and as such suffer from thermal inertia. That is, before they can flash, they must first heat up. They brighten slowly and dim slowly, whereas LEDs come on hard, stay on hard, and go off hard, in classic digital-step fashion. Such abrupt transitions are easier to see, extending the range of the lights. Meanwhile, LEDs keep on getting brighter. “I love it,” Green says. “Our raw material just keeps getting better. We don't change anything, just put in new LEDs, and this light that used to have a range of one mile is now a two-mile light.” In addition to which, the company is learning tricks to squeeze extra performance out of its LEDs. Lens makers are also adapting their designs to match the new technology. Carmanah's most powerful lights have a range of five miles.

Running a flashing light is hard on a filament—they burn out quickly. Carmanah's competitors had spent years developing special bulbs with skinny filaments down the middle, and lamp changers that, every time the filament burned out, automatically clicked over to another one. Such mechanical solutions were intrinsically expensive and required regular servicing. At one-tenth of the cost and weight, Carmanah's lights have revolutionized the navigation business. “We basically just turned the whole industry on its head,” Green says. He reckons that his company's penetration in navigation lights is only one-tenth of the potential market. “I mean, it still hasn't worked its way through the world, but it's happening. It's only a matter of time before all marine installations switch to LEDs.”

In the wake of their north-of-the-border counterparts, the US Coast Guard followed suit, making their first purchase of Carmanah navigation lights in March 2000. The coast guards were tough customers, insisting that the lights meet their rigorous standards, performing tests on them for years. Once satisfied, however, they bought in quantity. Their endorsement has provided wonderful cachet for convincing new customers.

In addition to its marine lights division, Carmanah also musters an aviation division that sells runway markers to airport operators. “They look a lot like the marine lights, but the optics are slightly different because they have to shine the light up.” One recent customer is the US military. “We've sold an awful lot to them for use in Iraq. The reason is they want to set up these bases, but they don't want to put in the permanent infrastructure because they don't plan on being there forever. You just put these lights down the side of the runway, and away you go. But we're also doing real airports in the US.” Here, too, the logic of LEDs is persuasive. They last longer, more than fifty thousand hours, versus three thousand hours for a halogen bulb. This ensures low-maintenance costs and better traffic flows, since airport managers are not constantly having to close taxiways to change burnt-out lamps. And LEDs use a fraction of the power, reducing an airport's energy bill (and greenhouse gas emissions). Estimates put savings for operators at 40 percent over ten years.

To meet Dave Green in Victoria, the pretty, pocket-sized capital of British Columbia, I take the clipper up from Seattle to Vancouver Island, a journey of almost three hours. As the boat noses its way into the little harbor, I note the incongruously massive state parliament buildings. These are fronted by a statue of the queen-empress and a towering wooden totem pole, the only indication of the presence of indigenous people, whom Canadians call “first nations.” At the far end of the harbor, the eye is drawn to an extraordinary asymmetrically shaped bridge. The light blue structure is dominated by a massive concrete counterweight at one end that enables the bridge to be raised and lowered in order to let masted boats pass underneath. Built in 1920, it was designed by Joseph Strauss, who would subsequently go on to design a somewhat more famous bridge just north of San Francisco.

I encounter Green at his office, above a cigar store across the way from the bridge. It is a cold, wet, windy, day in early April. “The good news is, this is about as bad as it gets,” he grins reassuringly. Green is tall and thin with longish windblown graying brown hair. Casually dressed, he looks every inch a yachtie. In fact, he comes from a nautical background: his father was a naval officer. It may just be because this is Canada, but his angular weather-beaten features remind me slightly of Neil Young. He has a lopsided smile and a shy, unassuming manner.

I mention the seeming mismatch of a high-tech start-up being located in Victoria (population: 335,000). This largely tourist town trades on its British Empire heritage, to the extent of even having a daily newspaper called the Times Colonist. “Victoria is actually a very dynamic place,” Green corrects me, “but in a small way, not a large way. You get lots of smart people here, because they want to live here and they're willing to move. There's a lot of innovation, artistic and intellectual energy, but it hasn't produced any large companies—it's a bit remote from the sources of finance—almost by choice, because I don't think people want to start something big. They just want to do their own thing. But under the surface, there's a lot of very clever small companies here, leading-edge firms.”

I ask Green what motivates him to start up companies. “I've always been really competitive,” he replies, “always wanted to accomplish things, but not focused around money. That's certainly not been the foundation for choosing to go here or there—you don't go into oceanography if you want to make money! So it's been more just following the interest, and the excitement of doing companies; y'know, when you're scientifically oriented or engineering oriented, it's very exciting creating things, building products, and it's great fun.” These days, however, he is more comfortable working behind the scenes as chairman of Carmanah. He has passed the role of CEO to Art Aylesworth who, coming from a sales background, is better equipped to function as the company's front man. This arrangement also gives Green more time to indulge in his passion, yachting.

Green shows me round the company's manufacturing facilities. These are located across the bridge from his office in a bunch of light blue weatherboard shacks that used to belong to the coast guard. The impression is of a large-scale cottage industry. But this is misleading because from this humble location, as Green proudly—and somewhat bemusedly—points out, they produce products worth millions of dollars a month. (Since my visit, the company has moved to a new manufacturing plant nearby.)

Initially, finding funding for Carmanah was hard. Local venture capitalists didn't want to hear about anyone who had made an actual product. “They wanted something Internet, or software oriented. I tried to tell them, Well there's software in this—it's got a very clever chip in it,” Green laughs. “Venture capitalists all follow each other. They always want to spread the risk, they want the other VCs to be enthusiastic as well, so they either all go, or none of them go. That tends to lead to this flocking mentality: Well, if he thinks it's good, I think it's good, too, and they all go charging off in one direction. There's certainly money around, but if you're off on the wrong tack, fighting the wind, it's hard.”

Initially, Green raised money by making presentations. Some angel investors, including one of his former partners, chipped in. “It was tough that first stage, but once we got rolling as a public company”—Carmanah was listed on the Vancouver Stock Exchange in 2001—“we found it quite straightforward to raise money.”

Navigation and aviation lights may just be niche markets, but Carmanah's core idea of coupling LEDs with solar cells also has plenty of other outdoor applications beyond buoys and runways. With the availability of high-power white LEDs, one example is bus shelters. The impetus for this application came from Ken Livingstone, the controversial mayor of London, himself an indefatigable traveler by public transport, who pledged to improve the service and safety of buses in the English capital. LEDs are used in bus stops and shelters in several ways. They illuminate the timetable, to let passengers know when the next bus is due. They also provide enough light to read by at night. Finally, they can signal the driver of an approaching bus that there is someone waiting to be picked up. Through the use of clever power-management techniques, Carmanah's solar shelters are designed to provide light even in the depths of London's notoriously dark and dreary winter.

The cost savings in terms of power of using LEDs in place of conventional lighting is around fifty dollars for one fixture per year. But the main source of revenue for bus shelters is actually the advertising panels they carry. Again, in order to be seen at night, these need to be lit. To fulfill this requirement, Carmanah acquired a small company based in (relatively) nearby Calgary, which specializes in edge-lit LED signage. In September 2005, after a design and development process lasting four years, Carmanah was awarded a contract to supply London with a minimum of fifteen hundred solar-powered, LED-illuminated bus stops. Other cities in the United Kingdom are planning to follow suit, as are, at least to some extent, cities in North America. However, as Green laments, “We're way behind—here in BC, we don't even have schedules on bus stops. The level we're at is, just to put a light there so that people feel safe enough to use the system at night.”

Having acquired the ability to light advertisements, Carmanah has moved on to illuminate street names and highway signage. For such applications, fluorescent lamps are conventionally used. These of course have to be replaced on a regular basis. Thus the logic of using LEDs here is the same as for traffic signals, a large and rapidly growing (100 percent per year in 2002) market that Green somehow managed to overlook. “I curse myself. I was focusing on navigation lighting and I didn't spot the traffic lights application. It just didn't occur to me.”

Possible compensations loom. For example, down the road apiece comes an even larger application: LED street lighting. Here, however, Carmanah will face stiff competition from much larger firms. For example, in Europe in 2005, Philips installed LED street lights in the Dutch town of Ede. The lamps are expected to last for fifty thousand hours, that is, eleven-plus hours a day for twelve years, four times longer than normal street lighting. Admittedly, these lights are not solar powered. But in Japan, Sharp, the world's largest manufacturer of solar cells, has commercialized stand-alone solar-powered LED street lights that require no overhead nor buried power cables, so they can be installed and turned on in a single day. They cost $4,500 each, which sounds like a lot, but it's much cheaper than digging up the sidewalk.

Dave Green, George Mueller, and Neal Hunter were all fresh out of college when they started their first companies. By contrast, when Don Evans started Cyberlux, he had basically retired and moved to the resort town of Pinehurst, North Carolina, intending to play golf. An economist by background, Evans had participated in starting up several successful ventures, mostly in medical electronics. Then, in 1998, it came to his attention that there was this gap in the marketplace. Evans just couldn't help himself; his entrepreneurial juices started to flow.

“What happened was, there was a study done by this agency,” Evans explained in his pronounced North Carolina drawl. “We're in an area along the Atlantic seaboard all the way down the Florida peninsula, then west into Texas and northeastern Mexico, that has a recurring storm phenomenon every year, from June the first to November the thirtieth, called the hurricane season. What is uniform throughout this horrendous experience is power outages. They had studied these power outages, and there was an incidence of, on average, 3.8 days without power that was directly attributable to these storms. The study asked individuals and business owners how they were affected by the loss of power in their homes and businesses. It wasn't in the script, but 80 percent of the people said, Why can't somebody come up with a light that will last for at least five to seven days on one set of batteries?

“Well, I happened to know the people that ran the survey, and they called me because they knew I was always interested in electronics. I went up to North Carolina State University, which is known for its very talented engineering staff, and I sat down and told the good professor there that I was looking for a light that was capable of illuminating a space twelve by twelve feet. I wasn't interested in a focused beam. I wanted to broadcast a blanket of light, but I wanted to do that for a period of forty-two hours on one battery set. He said, How did you arrive at forty-two hours? I told him, Well, the hurricane season occurs in the middle of daylight saving time, it really doesn't get dark until after seven o'clock, so if you had six hours of light from seven o'clock at night, that means you would be able to produce a good room-filling light for a period of six hours a night, for seven days, a full week. So he got on his blackboard, which ringed the room, and he started off here. He went all the way over there, he got to down the end, and he said, It just can't be done.”

But Evans is not the sort of person to take no for an answer. He called an old friend from his navy days, an MIT graduate in electrical engineering who had made a pile of money in fiber optics. “He told me, You need to look into the new white LEDs, that is wherein your solution lies. Resultantly, I went to work trying to identify engineering firms that were familiar with white LEDs. There weren't any. Then I found an industrial designer in Florida who had worked for BorgWarner, designing instrument panels for boats that he backlit with LEDs. But they were amber and red, that type of thing. Between the two of us, we got our hands on the Nichia white LEDs. They really provided the solution. And that was in 1999.”

Cyberlux, the company that Evans founded in May 2000, has not looked back since. Its first product was a multipurpose LED-based emergency light called EverOn. Designed to be hung from a picture hook on a wall, it contains six white diodes and four amber ones. They run on four AA batteries. How long the lamp lasts depends on how it is used. It has three settings: as a bright room-filling light, it lasts thirty hours; as a reading light, sixty hours; and as a night light, providing a diffuse but reassuring glow, a whopping five hundred hours. The company supplied fifty of these lights to relief workers clearing up after Hurricane Katrina in Biloxi, Mississippi. The workers were asked to return the lights, but none of them did. Evans was disposed to see this philosophically, as a pretty strong endorsement of the product. The EverOn is now available in retail stores. It reportedly sells particularly well during the hurricane season.

Cyberlux has subsequently branched out from emergency lighting into other market segments. They include LED accent lighting strips for under-cabinet applications in kitchens and bathrooms, as a replacement for compact fluorescents; also, for display cases in jewelry stores, as a replacement for halogens. (Retailers are sick of their staff burning themselves on hot halogen lights.) The company is based in Research Triangle Park, conveniently just down the road from Cree.

Obtaining funding for the start-up was, as always, a struggle. In the immediate aftermath of the dot-com bubble bursting, venture capital was well-nigh impossible to find. So, like the other LED entrepreneurs we have encountered, Evans initially approached angel investors. In August 2003 the company went public. Since then, Cyberlux has been able to use its stock to raise money. Around the time of the IPO, Evans brought in Mark Schmidt to take over day-to-day running of the firm. As a fifteen-year IBM veteran, it was natural that Schmidt should see similarities between the rapid evolution of the solid-state lighting industry and that of the personal computer. But there was also a big difference. “This time, we don't have to educate people on the value proposition around light, or energy efficiency, or not having to worry about maintenance,” Schmidt explained. “Everyone has that perspective. Everyone innately understands the difference between light and dark.”

Carmanah and Cyberlux are just two examples from the vanguard of a seemingly ever-swelling army of solid-state lighting start-ups. Others include BridgeLux, Dialight, Emissive Energy, Gallium Lighting, iLight Technologies, io Lighting, Lamina Ceramics, Lednium, Ledon Lighting, Lexedis, Lighthouse Technologies, Lighting Science Group, LightWedge, Luminus Devices, NuaLight, Optek, OptoLum, Phoseon Technology, SemiLEDs, Super Vision, and, as we shall see in part 4, Permlight. In addition, there are also dozens of Asian LED start-ups, most of them based in Taiwan, Hong Kong, and, recently, in mainland China. Almost all aim at providing lighting for the advanced markets of the developing and the developed world. In a sense, however, the most important application for solid-state lighting is in underdeveloped countries, to bring about a radical change in the lives of the poorest of the poor. One man has done more than anyone to demonstrate how to bring about this change. The next chapter tells his story.