James Montgomery, Associate Editor, RenewableEnergyWorld.com
December 12, 2013 | 1 Comments
New Hampshire, USA -- Are we seeing deja-vu in today's concentrated solar photovoltaic (CPV) market from the 2007-2008 period of solar PV? That's what one analyst says, offering as evidence wide technology advancements, a ramp-up in demand, and rapidly decreasing cost projections.
The PV market in 2007 and early 2008 was shaped by heavy technology development, and a lot of uncertainty as companies struggled to become "bankable" and get financing, explained Karl Melkonyan, solar PV analyst for IHS out of Germany. Even then, though, there were leaders in PV manufacturing, and in regional deployments — and then in 2008 the industry saw a "breakthrough" as demand surged and prices plummeted, he explained.
Similarly, CPV technology is comparably pushing forward very rapidly as PV did back then: in just the past few months we've seen recent records in cells (Soitec, Solar Junction) and modules (Semprius, Amonix). Improvements are also being made in the system optics (lenses), reducing the impact of higher temperatures from concentration, reducing transmissions through the system from the modules to the cabling, and optimizing the trackers.
In terms of uncertainty, we've seen that in the CPV market, not unlike solar PV in the late 2000s: SolFocus closed its doors this fall, GreenVolts lost ABB funding, Amonix closed its Las Vegas plant, Suncore acquired Zenith Solar, and Soitec closed a German plant to consolidate at its newer and larger San Diego facility. To achieve "bankable" status CPV companies are judged based on investors' confidence in their longevity of operation and installation of products and projects, Melkonyan said, and having insurance or a performance warranty helps too. (Note this fall Soitec began offering performance/warranty insurance for its CPV modules through Germany's Munich RE.)
CPV's Demand Trajectory
Demand for CPV by itself is lower than it was for PV in the late 2000s, of course. IHS sees CPV installations soaring over the next few years, from just 160 MW in 2013 to more than 1.3 GW in 2020, and more than doubling every year. More than half of projects assigned that account for IHS' 2014 and 2015 installation numbers, according to Melkonyan. Five vendors (Soitec, Suncore, Silex, Semprius, and Amonix) each will have hundreds of megawatts of installations by the end of 2015, he says, with a growing crop of startups and midsize firms coming on fast: MagPower, Morgan Solar, CSDR, SunPower, and Solaria, among others.
Global CPV system installation forecast. Credit: IHS
In the near-term there's still plenty of activity in CPV, though. Soitec is expanding into multi-megawatt CPV plants in South Africa and China, and has been busy with smaller pilot projects in Portugal, Saudi Arabia, and the U.S.. Suncor is working on its own similarly sized project. And SunPower is part of Apple's plans for a 18-20-MW CPV plant for its data center in Reno, Nevada.
LCOE vs. Cost Per Watt
CPV's installed cost-per-watt is rapidly decreasing thanks to improvements in system efficiencies and the economics of scaling up to volume business: Melkonyan says they'll decrease nearly 26 percent this year alone to $2.62/W, and are on a track for 15 percent annual reductions through 2017 to $1.59/W. That 2013 cost/W number is a global weighted average, Melkonyan points out; in China the lowest HCPV system cost will be $2.38/W, and prices are declining fast in South Africa and Middle East regions, with "hundreds of cumulative installations" by 2017, he noted. Those are CPV's sweetspot target markets: hot, dry climates with high direct normal irradiation (DNI greater than 6 kWh/m2).
Those cost/W numbers are admittedly higher than for PV — but CPV proponents have long argued to broaden the discussion to the levelized cost of electricity (LCOE), from generating electricity when it's connected and stretched to include total lifetime energy production and systems costs — and in those key high-DNI regions the calculations point to CPV success. By the LCOE metric, CPV's average LCOE is declining nearly 23 percent this year to $0.14/kWh and should decline another 12 percent annually until 2017, Melkonyan said. (Some back-of-napkin math puts that 2017 LCOE at around $0.08/W.)
Russ Kanjorski, VP of business development at Semprius, calls the IHS cost/W projections "realistic," and underscores that CPV "has the most headroom for efficiency improvement and system-level cost reduction over the next 10 years" — at least, he added, for the most competitive CPV variants. Amonix CEO Pat McCullough said he thinks IHS' pipeline and cost projections might even be on the low side.
The big wildcard for CPV in the next couple of years will be in the capacity/installation numbers, dependent upon what Kanjorski called micro-supply dynamics: market competition and shakeout, access to capital, commercial partnerships, and overall bankability. Total market penetration could vary significantly up or down from the IHS numbers depending on how those dynamics play out amongst CPV companies, he offered.
Lead image: Stereo VU meter, via Shutterstock