We had a strong response on this article and smart questions on our comment board. Dan Shugar, the CEO of Solaria, took some time to respond to the issues brought up by our readers.
***
Response from Dan Shugar
Thank you all for the insightful questions, which I will take a quick pass at answering:
Temperature. Great question, the first one I asked when I saw the module. It turns out our cells operate at the same temperature as normal silicon modules. Solaria built a number of modules with thermocouples inside of the cells, on the backsheet, and did the same for standard PV. Outdoor data under a wide variety of operating conditions shows temperatures within one degree C, sometimes more, sometimes less. This is because our strips are relatively narrow; the heat readily conducts and then radiates and convects to the air like a normal module. You can also simply feel the back with your hand, or look at it with an infrared camera, with the sample result.
Regarding the overall temperature of crystalline, it's true that modules get hot and performance drops off with temperature. And yes, sometimes in no wind situations, crystalline modules can reach 65 or even 70 deg C. However, on average, module temperatures are far lower, due to wind, which cools modules significantly, less-than-full sun, and other factors.. Manufacturers list NOCT, which stands for Normal Operating Cell Temperature, for each of the modules. Ours is 45 deg C, fairly typical for crystalline. Spraying water on operating PV does increase output, but is generally not a recommended practice in the industry. Keep in mind our modules are used on trackers, which are exposed to the wind, and run cooler than roof systems that are often mounted close to the roof deck.
Other questions dealt with independent performance measurements and bankability, which in my experience are closely related. We have indoor flash test data from four different labs that very closely matches our testing, which uses modules calibrated to national standards for reference. We have modules under outdoor testing at several national labs and more than six sites controlled by very sophisticated customers; for our Gen-1 product over 2 years, and for Gen-2 (current design) over a year. All data we have thus far received match our values, and we are conservatively rating our product for market introduction purposes.
Other questions dealt with capex. Yes our capex advantage is for the marginal capex. In the industry, there is roughly 10 gigawatts of crystalline capacity. If we converted all this cell material to Solaria modules, it would result in 20 to 30 gigawatts of modules, and our capex would be less than 10 percent of the industry average. Once that point was reached, then more upstream investment would be required, and our capex would be about 40 percent of industry average. This can be a confusing topic and I hope this explanation makes sense.
In addition to reducing the upstream polysilicon, ingot, wafering, and cell investments -- the "multiplicative" effect of Solaria technology greatly reduces footprint of PV manufacturing, reduces energy payback of c-Si to less than one year, and greatly accelerates that rate at which capacity can be added so we can replace that nasty coal power with clean PV juice.
Thank you for your interest in our company!
Dan Shugar
***
Original article follows...
Dan Shugar has been involved in the world of solar since the early days of 1988. He built PowerLight with Tom Dinwoodie (now CTO at SunPower), and that pioneering solar firm built over 500 large installations and grew strong from 1996 to 2006. SunPower purchased PowerLight in 2007 for $332 million as part of their downstream integration strategy.
Shugar worked at SunPower for a few years with strong results, taking a sabbatical in March of 2009. He then worked with the Sierra Club and helped "kill over 100 coal plants comprising 60 gigawatts of baseload coal."
Shugar also spent his time helping out some venture capital firms where he "evaluated dozens of technologies and employment opportunities."
Bottom line: He did not have a pressing need to return to work.
So when Suvi Sharma, the CEO of Solaria, a low-concentration photovoltaic (LCPV) panel startup, approached him in September of 2009 to consider a position with the company, Shugar initially said, "No thanks."
However, after some cajoling, Shugar and his own team of experts performed an extremely careful inspection of the company. And when everything checked out technically and financially, Shugar took the reigns of VC-funded Solaria as CEO in January of this year.
***
Disclaimer of sorts: I have been less than charitable to concentrating solar firms in my reporting. I have called the CPV business "a zero-billion dollar market," I may have displayed some schadenfreude over the demise of SV Solar, and I have labelled the majority of the herd of CPV startups as doomed. Oh, and I also intimated that Shugar's role as the Solaria CEO was akin to being installed as the new captain of the Titanic. (Shugar said that if he hadn't have vetted Solaria carefully, he would have thought the same thing.) And don't forget Sunrgi.
But a flexible mind is a sign of maturity or something, and I'm starting to see at least some light at the end of the CPV tunnel for a few firms. In high-concentration solar, Amonix just received an enormous funding infusion from Kleiner Perkins. They might have gotten it right. Soliant's investors tell me to keep an eye on rooftop CPV.
And during a recent call, I got tangled up in Dan Shugar's reality-distortion field and came out a lot less pessimistic about Solaria's LCPV technology.
***
Solaria just closed on a $45 million funding round led by CMEA Capital and DBL Investors along with Sigma Partners, NGEN Partners, Mitsui Ventures and Savitr Capital. According to Shugar, "The due diligence is an order of magnitude higher than it used to be."
Shugar spoke of "the dream of the solar concentrator": after 40 years of work, no one has developed a good product. When Shugar brought a Solaria sample to one of his mentors with 35 years of solar experience, "someone I hugely respect," this gentleman looked at the sample and asked, "Why did it take the industry so long to get to this product?"
Solaria’s technology is based on dicing or “singulating" a standard crystalline silicon wafer and mounting these strips on a substrate with a lensing system that essentially halves the requirement for silicon. The lensing and concentration is integrated into the rolled cover glass, representing a significant change from an earlier acrylic sub-assembly design.
"We have a new product," said the CEO.
I quizzed the CEO about the characteristics of the new product. It didn't take a lot of prodding, as Shugar recognizes the market sensitivities and he's not shy about speaking.
I asked about the yield of such a tricky operation as dicing these wafers. The yield is 99 percent and repeatable, according to Shugar. He added, "we lose no material" in the sawing operation and that the Bill of Materials (BOM) for the Solaria panel is about the same as the BOM of a standard flat panel. The sawing and assembly is performed on equipment that is well-proven from the semiconductor industry. And again, according to the CEO, the cutting of the silicon adds a negligible cost. "Our yield is over 99 percent -- we've done a gazillion strips, we've done it on a variety of wafer thicknesses....We own 40 megawatts of equipment now," he adds.
What about the size, weight and reliability of the panel? The glass is a bit thicker, which makes it heavier than a standard module but also more durable and rigid, according to Shugar. And because of the small size of the silicon strips and the impact of size on thermal stress issues, "If anything this thing is more reliable than a normal module." The panel is the same height and width as a standard panel with the same junction box. Shugar imported his own team of ex-Sunpower and ex-National Lab reliability experts to perform the reliability testing, prior to him taking the CEO role.
IP? Shugar has done a deep dive on the startup's intellectual property. Solaria has applied for 85 patents, with 30 granted.
Tracking and performance? The Solaria modules are designed specifically for ground-mounted tracking systems and certified to UL1703 and IEC61215 standards. In Shugar's words, "In any decent solar area that you deploy c-Si, it's a no-brainer to put it on a tracker." The energy yield of the Solaria panels was tested over a wide variety of meteorological conditions; the panels perform with direct radiance but also with diffuse light.
What about cost?
In terms of price per watt, "We reduce the silicon content and cost by half or two-thirds while adding glass cost, at scale, of a nickel per watt." Shugar expects Solaria's costs to be 40 percent less than the leading silicon manufacturers by next year.
According to Shugar, Solaria’s manufacturing processes require only a fraction of the capital expenditure per watt of manufacturing capacity needed by standard industry processes. In a market where cash is difficult to obtain, when Wall Street can't be relied upon for production capital, Shugar believes that capex is becoming more important than price-per-watt product cost.
This is how Shugar sees it breaking down: the big three in China -- Suntech, Trina and Yingli -- have a dollar-per-watt capex cost. High-efficiency crystalline silicon capex is on the order of $1.25 per watt.
Shugar makes his point by adding: "Our capex is under 20 cents per watt."
***
There are many firms in the CPV market -- too many, actually. We listed some of them in CPV Part One, Part Two and Part Three.