Your daily dose of Photovoltaic Technology Developments and Solar News
Report from Solar Power: XsunX follows own path to amorphous-silicon thin-film photovoltaic success
Most Popular Articles
- Top Stories 2008 – Power Generation – the good, the bad, and the efficiency
- Top stories 2008 – Materials – could UMG be the answer?
- Acciona’s Portuguese PV plant may be big, but one of the numbers just doesn’t add up
- Top Stories 2008 - Thin Film – orders, gigawatts and growth
- Top Stories 2008 – Cell Processing – efficiency is the name of the game!
Editor's Picks
Latest Videos
There are few places on the planet more suited to host a solar power industry event than San Diego, with its abundant sunshine and mild climate. But the attendees and exhibitors here for Solar Power International 2008 have little chance to enjoy the blue skies and warm breezes, since the real action's been taking place inside the bustling exhibit halls and conference rooms.
One amorphous-silicon thin-film PV start-up benefiting from the exhibit-hall bustle is XsunX. I spoke with CEO Tom Djokovich, who updated me on progress at the company's first 25-MW factory in Portland-adjacent Wood Village, OR. He told me the rehab of the building has been completed and a contractor has been selected. Since the facility was already built to semiconductor specs by the previous owners--with cleanroom air- and water-handling systems in place--there were minimal modifications needed to get the space ready.
Djokovich says the hook-up of the first manufacturing systems will begin in December, a slight pushback from the October time-frame previously stated. By April 2009, the line should be turned on, he said, and then the production ramp will proceed at a rapid clip. Depending on the speed of the ramp, he expects to produce between 6 and 12 MW of modules in calendar year '09.
XsunX's assumptions factor in 58% capacity utilization in the first year, which would result in an output of about 200,000 modules, according to the exec. The next step--80% utilization and yields--would produce approximately 333,000 modules, equivalent to about 35 MW (which exceeds the base capacity number), assuming the line is fully balanced and utilizing a workforce of as much as 160 to run the factory over three shifts. The targeted manufacturing costs are about $1.58 per watt for the initial line, with a roadmap to under $1.20 per watt.
The company's plans call for the build-out of an additional 100 MW in capacity by 2010, although the site of the next factory has not been selected. Djokovich indicated that if another Oregonian location could be found near the current factory, that would be "ideal," although no final decision has been made, noting "it will boil down to the financing options."
He estimates that with four more production lines, the cap ex will be about "$200 million, all in," and that although he anticipates that XsunX will be "cash-flow positive" by August or September of next year, it will need more money to underwrite its future expansion plans. He expects the company's efforts to close something on the capital formation front to push back to January or February, given the current financial crunch.
Unlike many other emerging a-Si TFPV companies that have chosen to go with turnkey production lines from the likes of Applied Materials or Oerlikon, XsunX is following what Djokovich calls its own "technology path" and selecting tools and materials from a variety of vendors. He claims that the company's $43 million in capital expenditures for its first fab amounts to about a third of the cost absorbed by competing companies that have gone the AMAT or other turnkey route.
For example, Praxair has been selected to help build out, hook up, and manage the gas delivery system. XsunX just announced that Spire will be providing some of the module assembly production tools, although not the complete Spi-Line 25TF turnkey line. Djokovich said the supplier will provide laminators, IV testers, and wiring tools, noting the wiring gear will be customized to XsunX's specific process requirements.
Other vendors include Shuttleworth for the material-handling and automation systems and Newport Specta-Physics for the laser-scribing equipment. As for XsunX's proprietary PECVD equipment design, he explained that work on the deposition process is focusing on keeping the interconnect and devices separate, calling them "grey areas." The CVD system will process eight panels per chamber every 12 minutes.
XsunX has been looking for ways to reduce the assembly time and materials costs in its module line and recently announced that it's looking at some alternative encapsulation approaches. The company has also opted to go with a frameless glass-on-glass design for its tandem-junction ASI-120 modules, believing it will help drive down the per-watt costs of packaging.
Djokovich showed me a recent study done for XsunX for Ibis Associates, which examined "the competitive cost position in thin-film module technologies in high temperature and low light conditions." Six competing PV products were compared--triple-junction a-Si, tandem-junction a-Si, microcrystalline-silicon, X-Si, CdTe, and CIGS--as well as two U.S. locations, Portland and Phoenix, using levelized cost of electricity (LCOE), NREL's solar advisory model, and other analysis methods.
Although the Ibis report's simulation model is only that--a model, not a data-set based on XsunX modules pumping electrons in the field--Djokovich has reason to smile at the results. Assuming a 1-MW, one-axis tracking installation in solarized Phoenix, with a module cost of $3/watt and stabilized efficiency of 7.9%, his company's a-Si outperformed the other five PV types, showing about 12 cents per kilowatt-hour in operational costs.
It also topped the competing panel flavors in terms of annual AC power output in the simulation. Amorphous tandem also fared well in the considerably less sunny conditions found in Portland, given its good performance in low and diffuse sunlight.
The results underscore Djokovich's contention that although conversion efficiency has its place (and XsunX's own technology development path includes higher efficiency microcrystalline and multiterminal devices), the most important metric is a PV system's ability to deliver kilowatt hours of AC power at the lowest cost.
By that measure, XsunX's modules could stack up quite well in the competition for solar farm and utility-scale applications--as long as the company executes on its production ramp, gets a reliable high-quality product to market next year, secures financing, and follows through on its longer-term growth plans.
