Solyndra, the thin film CIGS company in Fremont, Calif., remains a lighting rod for criticism. For some, Solyndra is a sign of the Obamapocalypse with control of our once-great nation ceded to socialists and a corrupt DOE in cahoots with evil venture capitalists. Others see it as a shining example of Silicon Valley innovation walking hand-in-hand with an enlightened DOE and an intelligent energy policy. More folks are in the former camp.
I had a long conversation with Brian Harrison, Solyndra's relatively new CEO, at their Fremont facility, and I tried to clarify the value proposition of the unique Solyndra product and get the firm's plans for the future.
If you read the papers or this website, you know the oft-maligned startup has taken almost one billion in venture capital, $535 million in a DOE loan guarantee, failed to make it through the IPO window and closed down their first factory amidst some layoffs. That's old news.
The future of the firm could take one of three paths:
- The end-game is a non-competitivesolarproduct based on a specious premise resulting in a $1.5 billion hemorrhage of VC and taxpayer money and a bankruptcy or fire-sale acquisition in a few years. This accompanies massive embarrassment for the Obama Administration, the DOE and a threat to the viability of a few VC funds, along with 1,000 newly unemployed Solyndrans.
- The company figures out a way to reduce product cost by half (they've already done that), then by half again, and then by half again in order to compete directly on an efficiency-adjusted basis with the Chinese crystalline silicon manufacturers and First Solar.
- The company reduces cost enough to be cash-flow positive in a large solar niche market and sticks with that premium flat, white C&I rooftop market. This future looks a bit like Germany's Sulfurcell, another CIGS company focused on commercial and residential rooftops, which is appealing to a large solar niche at a higher ASP as well.
While Solyndra works to lower their costs, there is a political food fight going on.
Fred Upton (R-Michigan), House Energy and Commerce Committee Chairman, has initiated an investigation into possible misuse of stimulus funds according to reports in The Hill. That would include the $535 million stimulus loan guarantee for Solyndra. The investigation is part of an effort by Republicans to probe the details of the stimulus funds to look for waste or fraud. The article quoted a GOP committee aide as saying, "We’ll be taking a close look at DOE as well as other stimulus funding within our jurisdiction."
Congressman Upton, in a letter dated February 17 to DOE Secretary Steven Chu, said that the travails at the high-profile startup “raise questions about whether Solyndra was the right candidate to receive a loan guarantee in excess of half a billion dollars.” Upton and Investigations Subcommittee Chairman Cliff Stearns (R-Florida), have demanded all emails and documentation between Solyndra and DOE regarding the loan guarantee be furnished to the committee within two weeks. The letter is reprinted below.
"It's a good project [that] we'll make work," Jonathan Silver, head of the DOE loan guarantee program, said in an interview with VentureWire last week in defense of the loan. He said that the challenges the company has faced are "growing pains" that accompany the development of startup technology companies. Silver went on to say that the loan guarantee program was tasked with backing innovative technologies, which involves inherent risks.
Amidst all the rancor is the fact that Solyndra has shipped a total of 100 megawatts, had revenues of $140 million in 2010 and is employing 1,000 Californians in one of the world's most automated and impressive solar panel factories. The CEO looks for the firm to be cash-flow positive by the end of 2011.
Solyndra did go through a DOE-induced recapitalization recently in order to continue to receive DOE funding and a $75 million debt facility, according to the VentureWire article. Several major equity investors converted their preferred equity into common stock and gave up their liquidation preferences.
On cost and efficiency
Harrison said, "From Q4 2009 to Q4 2010, we cut cost in half, and in the next year, we'll cut it almost in half again."
Back in December 2009, Greentech Media estimated from the S-1 filing that cost was about $6.29 per watt. So it looks like Solyndra can approach $1.50 per watt by the end of the year. That's not the $0.75 per watt that First Solar boasts, but as Solyndra would tell you, it's not an apples-to-apples comparison.
A February presentation by Ben Bierman, Solyndra's EVP, Operations and Engineering, calculates the efficiency of the Solyndra tube at 12.7 percent "if produced in flat plate form" with the best modules at 14.3 percent. The average panel has an output of 205 watts with a target of 250 watts by 2013 according to the presentation. First Solar panels were at 11.6 percent efficiency in the most recent quarter at about $0.75 per watt.
Production scaling and DPPM
The firm is processing 2 million tubes per month and is on the way to 50 megawatts of production per quarter. Bierman's presentation claims a total of 107 defects out of 400,000 panels shipped, or 273 ppm.
Fab 2 Versus Fab 1
Fab 1 was not originally built to be a large-scale PV module factory and was constructed "like the Winchester Mystery House" with posts in the middle of the factory floor interfering with equipment and product flow. Harrison cited an example where three moly sputter tools were in three different rooms operated by three people. Fab 2 is now an "ideal, efficient layout" with a significant labor savings -- the same number of people can operate the 100 megawatt Fab 1 as the 300 megawatt Fab 2, according to the CEO.
Manufacturing in the U.S. Versus Abroad (China)?
"We're proud to be made in America and that we're manufacturing here," said the CEO, acknowledging, "You're never going to compete [with China] on labor cost. We can compete and win on innovation." He added, "Low load, low balance-of-system are the primary benefits on a value-engineered roof. We will not win on cost in a pure commodity market and don’t try to compete on purely commodity priced deals. We target a specific market with an efficient system level solution."
(While we're on this topic, read Shyam Mehta's piece on solar manufacturing in the U.S.)
Solyndra plans to stay in the commercial and industrial roofing market, estimating it as five to six gigawatts. The firm has introduced a new product intended for greenhouses -- a surprisingly large market and one suited for Solyndra's form factor.
As for the location of the next factory, Solyndra has the option to build a mirror image of Fab 2 next door, but that will depend on where the market growth lies. The CEO emphasized that "people care about buying American."
The Importance of Saving on Balance-of-Systems, Ease of Installation, and a Light Load
When you buy a panel from Solyndra for, say, $2.30 per watt, the mounting and all of the internal wiring are included. There is no junction box. The only thing the installer or integrator has to deal with are the home runs and the inverter. There are also far fewer trips up a ladder that are necessary, according to the firm. Solyndra also claims that because of the shape of the power output, the system can use a lower peak power inverter and save money on the electronics. According to the CEO, if the customer looks at BoS costs, "we rarely would lose on price; if somebody looks at panel cost per watt, we will always lose." Bierman's presentation claims that Solyndra is "always competitive with crystalline silicon" when judged on an "all-in" basis.
Note that the module is non-penetrating and requires no tools to install. The cylindrical design provides a 1.5 concentration when mounted on a white roof and also makes the module more tolerant to being oriented to roof perimeter lines.
Jamie Hahn, managing director of of Manasquan, N.J.-based Solis Partners, a solar developer and integrator, has used panels from Solyndra on some of the company's projects. Hahn attests that "Solyndra is good for high wind" areas and roofs without a lot of reserve weight-bearing margin. New Jersey is the Saudi Arabia of flat roofs, and Hahn estimates that only about 60 percent of the roofs can accommodate the additional 6 to 8 pounds per square foot that a ballasted c-Si panel installation would add. Hahn also mentioned to us last year that building owners can claim the investment tax credit for a new roof when using Solyndra tubes.
Harrison said, "Our company has had growing pains, like many Silicon Valley startups transitioning to full production. But we have moved forward and are proud to manufacture here in Fremont."
The takeaway and the value proposition for Soyndra is that their innovations include panels and mounts, and simplifies the cost and work involved in getting lightweight, non-penetrating solar on flat roofs. Despite the innovations and differentiation -- Solyndra still has to scrub out cost in a big way in a market with plunging module and BoS costs. Perhaps the 1,000 beleaguered entrepreneurs in Fremont can rise to the occasion.
Here is a partial list of Solyndra installations:
- 1.2-megawatt rooftop in France
- 370 kw in Italy
- 1.28 MW in France
- 1 MW in Germany
- 1 MW on Frito-Lay plant in Modesto, Calif.
- 1.23 megawatts in the Czech Republic
- 4180 panel at PSE&G in New Jersey
- Capital costs below $1 per watt on the first 100 megawatt line (and declining from there), and a production platform that leverages proven equipment.
- Panels made using larger (than FSLR) monolithically integrated circuits as opposed to small unit cells, which generally have higher costs as well as significant manufacturability and performance issues.
- Designs that drive installation and BOS costs lower.
- Advantages that are particularly valued by a substantial market; e.g., a uniform, all-black panel with superior aesthetics makes a big difference in the residential market. (Likewise, pay attention to potential disadvantages for markets.)
- Attributes that exclude others from a substantial market; e.g., completely cadmium free / lead free product is critical for some geographies. (Likewise, pay attention to attributes that may outright ban it from markets, now or in the future.)
- DOE loan guarantees are usually a good thing, but only if the company has an economic product and is ready to scale. I [Khosla] would not interpret a DOE loan guarantee as “validation” of costs or a reason to IPO. Would you rely on somebody in government to understand the dynamics of competitive costs in the global marketplace? Nor would I rely on bankers offering IPOs to have validated a technology either, especially if they are getting commissions for the deal. Also remember that a big loan gets the treadmill running and cash flowing out to repayment of the loan. Great product margins are the only source of cash for paying off a big loan (without dilutive equity raises).