Search sponsor
Search sponsor
CIGS thin-film PV production, though still in its infancy volumewise, is rocking and ramping at an ever-faster clip. Solar Frontier (ex-Showa Shell) claims that it will have more than 1GW of CIS capacity online by 2011. Q-Cells says its Solibro unit more than doubled its production over the last quarter to nearly 9MW, as the German copper-indium-gallium-(di)selenide manufacturer remains on track to increase production output to 80MW on its two lines by year’s end. Professing 10.5% average conversion efficiencies, Ascent Solar touts the official commencement of production on its Fab 2 line in Colorado, strengthening its position as the only company manufacturing monolithically integrated flexible CIGS modules using plastic substrates.
Although Ascent has a cool virtual video tour of its new production facility, there’s nothing like a old-fashioned factory visit to put things in perspective. I had a chance recently to walk the bustling manufacturing floors of two privately held Silicon Valley-based players—MiaSolé and SoloPower—a pair of companies that should be included in any conversation about CIGS pushing into production mode.
The two firms dwell at opposite ends of the South Bay corridor. MiaSolé’s two facilities, within short driving distance of each other, are tucked among the scores of other nondescript tech buildings in Santa Clara. SoloPower’s HQ resides in the more spacious, quasi-bucolic Edendale tech park off Highway 101 at the extreme south end of San Jose, not far from fellow CIGSters, Nanosolar and Stion (yes, the word on the street is that Stion is indeed working on some variation of the quaternary PV cocktail).
Both companies, recipients of hundreds of millions of funding between them, have taken their lumps from skeptics and CIGS naysayers as they have progressed from stealthy start-up to early stage and now into pilot and volume production. The criticism once had some merit, since the hype around the two outfits far exceeded any real (or imagined) production output.
Management changes, first at MiaSolé, then more recently at SoloPower, seem to have put firmer hands on the tillers of each solar ship, with a new-found commitment to optimizing efficiencies, commercializing the product, ramping manufacturing, and shipping modules. No more cash-burning, science-project dog days for these guys.
Although the two companies share a propensity for roll-to-roll processing of their CIGS cells on flexible metal foils and the unnecessary capitalization of a letter in the middle of their names, MiaSolé is admittedly a fair piece farther down the commercial road(map) than SoloPower.
The latter firm’s advanced R&D/pilot line may be fully operational and impressive in its own right (as the above photo attests), but it is only processing limited amounts of flexible CIGS foil from its electroplating-based process line and producing small batches of flexible modules for shipment to customers.
New president/CEO Tim Harris said they will be ordering high-volume production tools soon, for what they expect to be the first 75MW (with efficiency improvement headroom to 100MW) line. He’s as excited about the high-quality team that’s been assembled—lots of ex-IBMers and the like—as he is about the investments that the company has made in areas such as automation, barrier layers, and module frontsheets really that are starting to pay off.
The starting gun signaling the next phase for SoloPower—commercial-scale manufacturing—will go off when the results of module certification testing comes back from UL early this summer. Harris and executive chairman Lou DiNardo exuded confidence that the modules would run the gauntlet with flying colors, since they’ve endured tests at least as challenging as the UL battery on the company's in-house gear.
The other news they’re anxiously awaiting is the decision on a Department of Energy loan guarantee application, which the two are also confident will be made in SoloPower’s favor. Soon, they believe, many of those letters of intent from customers should convert into supply deals.
Up the highway at MiaSolé, CEO Joe Laia smiled about the progress they’ve been making: “we’re stoked,” he told me in true Californese. “We can really make modules, we ship them, and you can go find them installed in the field.” And the corporate “we” includes over 300 employees these days.
The company’s fully certified modules have been shipping for revenue since October 2009, with expectations of getting at least 5.6MW of product to customers like Phoenix Solar in the first half of the year and a few mutiples of that for 2010 as a whole. Systems with MiaSolé-equipped arrays have been built, are being built, or are in the development pipeline in Europe, North America, and India.
Conversion efficiencies continue to head north, with NREL tests soon expected to confirm a significant jump in the median module efficiency coming off the line, currently at an “official” level of 10.5%.
“We know the number, and it’s much better than we ship today,” according to Laia. “If [efficiency] is not scaleable, it’s not helpful from a business standpoint.”
Dozens and dozens of boxes packed with incoming glass or finished ~112W-rated frameless modules were stacked on the floor when I had the guided tour of what the CEO calls the “100MW prototype production facility,” a 92,000-sq-ft site incorporating both the company’s second-generation roll-to-roll CIGS sputtering systems and fully automated (and I mean, fully) moduling lines.
The factory was turned on in May 2009, went through debugging during the summer, and has been outputting modules incorporating two rows of 44 cleverly interconnected CIGS cells each since Q309. The layout is impressive, a best-of-breed plant sporting both customized and off-the-shelf gear on a par with any solar production house.
Two 20MW coaters remain at the company’s main building, one which rolls out cells for production and another earmarked for R&D efforts. At this point, the “back-end” module assembly capacity outstrips the “front-end” cell-strip making capacity (although the mods squad still needs a few more laminators and lay-up stations), but the gameplan calls for a new 20MW coater (which looks like a proper piece of semiconductor process gear, when compared to the first-gen models) to come online each quarter, according to Laia.
As he explained, a sputtering platform is released to the process guys who “get a quarter to show what it can do” before the tool’s qualified and handed to production. Once it goes live another tool is readied for its process qualification paces while a third one begins to be built up.
Since we arrived at the plant as the shift was drawing to a close, the robots were not schlepping stuff and whipping around as much as they might have earlier in the day, and there were engineers performing manual inspection measurements here and there.
The boss said that although plans call for 24/7 operations with four shifts and eventually “no human [will] touch the module until the end of the line… we don’t run the factory seven days a week today. We toggle between production and engineering. We have a whole bunch of engineering projects that we’re working to close, to give them a chance to put what they’ve done on the line and then hand it back to production.”
When I inquired about capital expenditures for the factory, Laia turned to CFO Merle McClendon and asked, “can we say?” and was given special dispensation to talk about that aspect of compensation.
“By the end of 2010, we will see 50 cents per watt capex in this facility,” a figure that works out to about $10 million for every 20MW, give or take a few percentage points.
He did admit they had encountered a “few gozintas” (as in “goes into,” and its corollary “gozoutas”) that had added a little extra unforeseen cost. “The first time you do anything, almost by definition, you don’t get it right.”
Speaking of costs, the bluejean-clad, dog-loving exec offered both praise for a certain thin-film company and a withering blast for those firms who don’t get to sub-buck-a-watt levels in a hurry.
When First Solar’s name came up in regards to low-cost manufacturing and how the leading thin-film firm has paved the way for next-wavers like MiaSolé, Laia enthused, “they really rock, they’re our idols.”
But as for the TFPV wannabes, he excoriated, “if you can’t see a way to get your costs down to 85 cents within the next 12 months, then I don’t see how you’re gonna be a player, how you’re gonna survive.”
It’s been said before but bears repeating: only a handful of hardy CIGS companies will likely survive and thrive, either through M&A or IPO, with the winners emerging over the next few years. Whether it’s a year, as Laia posits, or a bit longer, that window will soon be closing and the amount of defenestrated thin-film PV players may far exceed those still in the race.
PHOTOS COURTESY OF SOLOPOWER AND MIASOLE
