There were three of us out there in the field, looking up at the California sky. All around was a landscape of scrub brush and wild grass, bending in a hard wind and stretching for mile after empty mile toward a range of snowcapped peaks.
The nearest dirt road was a half-mile away; it ran crooked and confusing for a few rutted miles to a paved local thoroughfare, which in turn led to a narrow highway and the hour-and-a-half drive back to Los Angeles. The only sign of civilization I could see was a nearby house where an off-the-grid family had amassed a junkyard's worth of car parts. Also, there was the odd thing we were looking up at: a circular, 155-foot-high wind turbine, built by a Massachusetts company called FloDesign and long rumored to be a major clean-energy breakthrough. As a design prototype, the turbine was one of a kind. Yet the expectation among the FloDesign employees taking me around the site was that next year it would be joined in California by nearly a thousand almost exactly like it. In other words, after a long and difficult gestation period—years of work, done almost entirely in secret, funded mainly by a host of California venture capitalists, the U.S. Department of Energy, and Goldman Sachs—this peculiar technology seemed finally ready to scale up.
FloDesign, soon to be renamed Ogin, is a vestige of a widely disavowed chapter in Silicon Valley history. Around the time of FloDesign's birth in 2008, one of the heated debates in the Valley was whether the next Google would arise in social media or in clean energy, an investment class that was gaining favor at several VC firms around the Valley, especially at Kleiner Perkins Caufield & Byers. An early Google investor, Kleiner plowed hundreds of millions of investor dollars into about 40 clean-energy startups, including FloDesign, making everything from biofuels to solar panels to electric cars.
We know how that contest turned out. Social media exploded; most of the clean-energy startups crashed and burned. FloDesign has so far survived. But its experiences over the past few years show that compared to change in the digital realm, with its billion-dollar startups that purportedly change the world over the course of a TED Talk, the job of making a transformation in the physical realm—of inventing a new technology; of finding an opening in the market; of devising a business plan that works; of fomenting the replacement of entrenched systems, in this case no less than our means of producing and consuming power since the electrical grid was assembled nearly a century ago—remains agonizingly hard, and the payoff long in coming. Kleiner invested $12.5 million in Google in 1999; Google went public in 2004, making Kleiner's stake worth $2 billion. By contrast, in year five FloDesign is just getting ready to launch its first product.
As FloDesign went through a variety of ups and downs over the past few years—delays stemming from storm damage on a test turbine to a redesign of its hardware—I kept tabs on its progress, meeting on occasion in Massachusetts and New York with its executives and engineers, who tried to explain why creating clean-tech companies was so difficult. "These are not Internet startups," Bill Joy, a Kleiner partner who has overseen the FloDesign startup and sits on the company's board, told me recently. "This is real stuff. There was a lot to learn," from mastering the aerodynamics to perfecting the design to creating from scratch an entire manufacturing process and global supply chain. Jantoon Reigersman, a company executive who accompanied me to the California installation, called the problem "a two-hump camel." First you have to prove that you have a radical new technology; then you have to prove your bankability to financiers and markets, or you'll never acquire the money to scale up. Getting the turbine to the California demonstration project and signing on clients for the next two years only mean the technology has made it through the development process to the equally fraught commercialization phase.
Normally FloDesign would buck up against some powerful incumbents—the makers of conventional wind turbines, including Vestas of Denmark, the world's largest turbine manufacturer, and General Electric. From the start, FloDesign concluded that it wouldn't compete directly against them. Instead, it would try to challenge the assumptions on which the wind industry is based. Rather than building enormous turbines at remote locations, it would build wind turbines that are small, powerful, and sited locally—near residential neighborhoods, for instance, or on industrial campuses. Doing so would mean that the company would have to create not only an innovative new product but also an entirely new market. As we looked up at the FloDesign turbine in south-central California, Reigersman said to me, over the whoosh of a steady breeze, "This isn't just about making a new wind turbine. It's about changing the way we think about wind."
If you had to choose one clean-energy technology that has already proven itself to be revolutionary, it would have to be the windmill, the current incarnation of which is known in the jargon of the business as the three-blade, upwind turbine. In the United States alone, there are now about 45,000 such machines, spread through 39 states, and thanks to sophisticated methods for handling what is known as intermittency—the wind isn't always blowing—their variable electricity output is now seamlessly plugged into the power grid. Government incentives have helped the wind industry along over the past few decades, mainly in the form of tax credits. But at this point, turbines are a legitimate triumph of economics, too, and can be cheaper in some instances than either coal or gas plants, both of which produce substantial amounts of carbon dioxide. Thanks to wind power's efficiency, in fact, last year was its most successful ever, with national capacity reaching 60,000 megawatts, enough to power the equivalent of about 15 million U.S. homes.
As the wind industry has grown, its turbines have gotten bigger too—so large that a few offshore turbines now top 500 feet in height and cover a 400-foot diameter in the sweep of their rotor blades. "Larger and larger and larger is the trend," Fort Felker, the head of the wind technology center at the National Renewable Energy Laboratory (NREL), in Boulder, tells me. Felker is not sure if we know yet the upper limits of turbine size, but he notes that the largest ones, placed in strategically windy spots (Texas, North Dakota, Great Britain's North Sea) can generate huge amounts of electricity at extremely low costs. Size can have some drawbacks: Big turbines are a challenge to engineer, build, transport, and install. They can also be too noisy and imposing to go in populated areas. (If you've ever stood below a big turbine in a steady wind, the autonomic fear response you experience at the sight of the huge, arcing blades bearing down on you is akin to how you feel when you get too close to the tracks of an onrushing train.) Still, the cheap, clean energy they produce easily outweighs all other factors and likewise suggests they would be a key part of any low-carbon (or zero-carbon) future.
FloDesign effectively began as a theory. In the sleepy town of Wilbraham, Massachusetts, two semiretired aerospace scientists named Walter Presz and Michael Werle had set up a skunk works for exploring jet-engine propulsion technologies. When I met with them there several years ago, Presz told me that around 2004 the two started to think: Rather than putting energy in for propulsion, what if you took energy out and turned the engine into a wind turbine? The men contemplated placing a shroud—a type of circular fiberglass frame—around the fan blades. The idea was an old one, but Presz and Werle rejiggered their shroud (they call it a mixer-ejector) to accelerate the flow of air through the blades. "It's a pump without moving parts," explained Presz. The upshot, he and Werle claimed, was a large boost in performance. "In our machine, a 2-mile-per-hour wind will be like a 4-mile-per-hour wind," Presz told me. "A 10-mile-per-hour wind will look like a 20-mile-per-hour wind."
FloDesign got off the ground in 2008, when Presz and Werle built a small scale model and tested it in a wind tunnel at MIT. "The surprise was that it validated our expectations," said Werle. Other people were surprised as well. Soon the company was winning clean-energy competition prizes; soon, too, a procession of venture capitalists and wing-tipped bankers began coming to Wilbraham, calling at the door of Presz and Werle as their limousines idled in the building's muddy parking lot.
As FloDesign developed the technology over the next five years, the company decided to go into stealth mode, refusing to publish data about its tests and power production. Its silence coincided with an investment by the U.S. Department of Energy's Advanced Research Projects Agency-Energy (ARPA-E), which helped fund the turbine's development with an $8.3 million grant. In the view of ARPA-E, pushing performance data into the public sphere too soon might attract foreign competitors and threaten the prospects of a business being helped along by U.S. taxpayers.
Yet the secrecy contributed to controversy about FloDesign's technology and stoked dismissive comments. When the company began introducing itself to the public around 2008, claims by its engineers that the turbine could double or triple the efficiency of conventional wind turbines elicited doubts within the industry's engineering community. Some argued that the upstart company was asserting results that contradicted established laws of physics. (In particular, they argued that the turbine could not, as claimed, surpass something known as the Betz Limit, which describes the maximum efficiency for conventional wind turbines.) Mark Johnson, a program director at ARPA-E, maintains that FloDesign's technology does not bend any scientific rules. "There was no new fundamental physics that had to be discovered on this," he tells me, adding that his group has been "very, very pleased" by how the company has met its technical milestones over the past four years.
To get ready for the market, FloDesign went far afield for its CEO. In 2009, Lars Andersen, a native of Denmark, was running the China wind business for Vestas. After he rebuffed a headhunter who asked him if he might be interested in running FloDesign, Bill Joy tracked him down at his base in Hohhot, the capital of Inner Mongolia—"an incredibly windy, desolate, remote city," Joy recalls—to persuade Andersen to take a deeper look at the technology. "I was skeptical," Andersen admits, "so I did as much research on the concept as I could. And then I [went] to see the founders, Dr. Presz and Dr. Werle, and I locked them in a room all day and asked them questions." He recalls thinking: "If this is really nearly as good as they are saying it is, it has the potential for a breakthrough." Today, Andersen is convinced that because of its inherent efficiency, the FloDesign turbine could generate electricity at relatively small scales—say, 100 kilowatts, which is enough to power a large building or 30 homes—for the same unit cost as today's big gas, coal, and wind installations, which might be rated at 100 megawatts. For the first time, Andersen claims, small-scale applications could enjoy the economics of large-scale plants. He says: "No other technology can do this."
In the energy industry, having an intriguing product should never be confused with having a winning business plan. No wind-energy company has ever succeeded in commercializing a shrouded turbine. You can chalk up the failure mainly to economic factors. The most important measure of any new clean-energy technology is how much electricity it can produce and at what cost. And if you add a shroud around the fan blades—a significant amount of additional materials and construction—you're adding a substantial cost. As Felker of the NREL notes, you might instead make the rotor blades of a conventional turbine 20% or 30% longer, a more straightforward method for producing more energy (and one reason why turbines keep getting bigger). Felker likewise remarks that shrouded turbines can also be more vulnerable in stormy conditions. FloDesign learned this lesson firsthand in 2011 when Hurricane Irene disabled a test turbine in Boston Harbor. By contrast, conventional wind turbines will "feather" their blades so as not to be damaged by high winds.
Under Andersen, FloDesign is betting it can lower its production costs and resolve any storm vulnerability; it has flaps on the shroud that can close down in a storm. But to capitalize on its claims to superior aerodynamics, the company decided to focus on a market that the big-wind turbines don't, and can't, serve. Less-windy sites and more environmentally sensitive areas are obvious places to start. FloDesign will also look to put its turbines alongside highways, or within island communities, or on military bases. This model is known as distributed energy, placing the source of energy generation close to the load—that is, near the user. Local energy sources reduce the reliance on the power grid and on big, centralized power plants. If the circumstances are right, distributed power can save much money and eliminate a lot of carbon emissions.
The distributed model has long since left the realm of tree-hugger fantasies. It has grown steadily in popularity over the past decade thanks to a variety of technologies, most notably solar power. What is more, because wind and solar are so compatible—turbines are often most productive at night, when solar is not, and vice versa—in theory the two could be installed together.
Yet distributed power has never caught on in any meaningful way with wind, and for good reason: So-called small-wind turbines—human-scale, 2- or 5-kilowatt machines that might help power a single home—are too expensive and inefficient. With today's industrial-size turbines obviously too big for such settings, you could sum up FloDesign's gambit as a Goldilocks strategy: Go medium. Build a turbine that is not too small and not too large, and turbocharge the output by incorporating what Werle and Presz, and Joy, too, believe is a significant design breakthrough. If it catches on—a big if, which depends largely on whether the price of its electricity is low enough—the FloDesign turbine really could be revolutionary.
Finding the sweet spot in the middle can give the company a couple of advantages as it tries to become a legitimate business. The first is that sticking to a maximum size of 100 kilowatts qualifies FloDesign until 2016 for a special federal tax credit for small turbines. The second is that producing turbines of modest size could help the company scale up before distributed power catches on by letting it compete in the U.S. repowering market—the replacement of thousands of aging wind turbines that are now reaching the end of their lifetimes.
Over the past 30 years, three of the windiest areas of California—Tehachapi, Altamont, and Palm Springs—have been host to huge wind developments that pump electricity into the state's power grid. Managed by a variety of companies, these wind farms—dense clusters of machines set in vast, isolated stretches of breathtaking terrain, far from the state's population centers—have long been an essential part of California's landscape as well as its economy. Early on, FloDesign's engineers tried to calculate whether their machine could serve as a replacement turbine in these wind farms. It might look like an alien invader, but its size is right: The old turbines are small, and FloDesign's turbine seemed geometrically compatible, which could let the company avoid environmental permitting hassles and the technical challenges of integrating larger turbines on sites created for smaller ones. (Overly big turbines can steal wind from smaller ones nearby.) "Nobody knew how large this market was," one person at FloDesign told me, "so we literally walked every site in California." The engineers counted 25,000 turbines in the state. The company concluded that if it could succeed as a California repowering company, it could survive as it scales up and creates its niche in the distributed-energy market.
The commercial process of getting the turbines on the ground, though, is, as FloDesign's Reigersman tells me, incredibly complex. We are driving back in a rented minivan from a visit to the prototype turbine outside L.A., and Reigersman, who serves as CEO Andersen's aide-de-camp, is describing what the company's next few years look like. Before coming to FloDesign, Reigersman worked at Goldman Sachs, and if anyone at the company understands the infinite variations of risk, it is he—apart from his work in banking, for fun he pursues world records in mountain ascents and deep-sea diving. FloDesign is currently facing numerous certification reviews for safety and performance; it is negotiating complex agreements with wind-energy developers and utilities; it is learning to manage a manufacturing supply chain that begins in China and ends in the small California town of Adelanto at a former yacht factory that the company acquired to assemble the turbines. In the meantime, Reigersman says, FloDesign engineers in Massachusetts are tweaking the aerodynamics while engineers in Denmark are perfecting the software that oversees and regulates the turbines' functions. The bottom line is that the company will install 10 turbines at the end of this year on the site of its first customer, in California's Tehachapi region, and about 850 total next year. To help the rollout move faster, FloDesign will also create an energy production company to build wind farms and sell power. Reigersman predicts the company will be cash-flow positive by the end of 2014.
And yet . . . Andersen sounds frustrated. "It has taken longer than I thought, to be brutally honest," he tells me a few months later. Negotiating with utilities has been trying, albeit unsurprising. At the heart of the energy market is a contradiction: a desire for both innovation and stability. "Everyone wants to be customer 4 or 5," he says, only half-joking. "Nobody wants to be customers 1 and 2."
To answer the question of whether a new kind of enterprise is likely to succeed or fail, people tend to hunt for analogies. In discussing his venture firm's investment in FloDesign's technology, Kleiner's Joy favors the analogy of computing. Big-wind turbines are like the old mainframes that large groups of engineers would share to make their calculations. Smaller turbines on the FloDesign model are like the PC. "We're trying to make it available to everyone," Joy says. "You can stick these turbines everywhere." He acknowledges that establishing the technology will take some time, but not forever. "If we're going to become sustainable, we have to reinvent," he says. "And the kind of ventures that can truly reinvent can take nearly a decade."
The comparison is helpful, to a point. It speaks to how the stakes in clean energy are enormous, even if the potential social value is not yet reflected in the current financial value. What the analogy does not capture is the possibility that a big energy innovation can take far longer than a decade. So far, it has taken about 100 years of incremental improvements to get conventional wind turbines to this point—plus prior improvements in Europe and Asia dating back 2,000 years, when wind currents were first harnessed by machines to pump water and mill grain. Watching FloDesign mature over the past few years, I could only conclude that the task at hand—making the technology and business succeed on an accelerated time scale—is grinding, difficult, and expensive work. Also, failure is always an option. "We have great optimism and great hope for them," says ARPA-E's Mark Johnson. "But it's like watching your children go to college." His point is that even though the company has overcome some risks, predicting what will happen next is impossible.
FloDesign's engineers and executives are hopeful that they will be an exception to the clean-tech class of '08. They base those expectations on how the right technology can combine with the right policy to produce the right economics and the right business plan. They likewise perceive this moment in time as unique, as climate change increases the urgency of a transition to a low-carbon, or carbon-free, world. "The governor of California wants to bring 12,000 megawatts of distributed power to [the state]," Andersen tells me one day in New York. He's correct, but it isn't all going to be wind; a lot of that power would no doubt be rooftop solar panels, or home heating units that can use natural gas to produce electricity. Andersen nonetheless seems assured that if he peers beyond the repowering venture in California, the world is moving in the right direction for his company. Two years ago, he remarks, he commissioned a study by an independent consulting firm to gauge how many "community sites"—that is, distributed generation sites—existed in the United States where FloDesign could profitably deploy its medium-size turbine. Later, I look over the report for myself. Using data culled from the NREL, the consultant calculated the FloDesign turbine's specs; then, it spit out an answer. The turbine could have "positive economic returns" on sites covering 35,796 square miles, meaning the United States alone holds the potential for about 3.5 million distributed turbines like FloDesign's.
That's a lot of turbines and a lot of money. Andersen sees the turbines spreading even farther, to Japan and Europe, too. Yet it is just a hypothesis; if FloDesign's business plan falls short in the next few years, the company's fate would be all the more bitter, since the potential seems so large. People love the idea of a radical new clean-energy technology. But actually taking a chance on one is another thing altogether.
[Photographs by Bob O’Connor; Illustration by Fast Company]