Solyndra Seeds Doubts About Photovoltaic Manufacturability

Ask a venture capitalist about the collapsing green-energy sector, exemplified in the successive bankruptcies of Solyndra and Beacon Power, as well as the downgrading of First Solar, and you’ll hear long soliloquies about the collapsing popularity of green-tech/clean-tech initiatives, and the failure to develop sound business plans.

Ask a Libertarian politician the same thing, and be prepared to hear about the inherent evil of federal subsidies and loan guarantees. Ask a globalist, and the person with Asian expertise is bound to tell you how the ability of China to emulate manufacturing processes has doomed the US alternative energy market.

Solyndra Seeds Doubts About Photovoltaic Manufacturability

Thin-film photovoltaic modules from First Solar.

These answers all are correct, in part. But ask a designer with a background in either civil engineering or energy electronics, and you’ll also hear that much of the boom in energy startups in the past decade has focused too excessively on unique point solutions to problems of solar conversion, point-source storage of transient energy such as wind-farm energy, and other proprietary inventions that look great in demos, but seem to exist in a vacuum. In fact, CNet blogger Martin LaMonica says the problem of scaling up is so ubiquitous, it may even affect the electric vehicle industry.

In one sense, you could say that the Department of Energy’s smart grid initiative has been one of the only alternative energy efforts that look to large-scale systems first, but even the smart grid software projects fail to take into account the antiquated state of the nation’s energy distribution network. Point solutions for residence or utility companies often fail on several levels. Can the proprietary design really be manufactured in a cost-effective manner? Will its ergonomics fit in known energy subsystems? Can the concept be emulated easily by an Asian company with lower costs of manufacturing?

Startups developing products like solar PV panels or radical flywheels could have stolen practicality lessons from the hardware startups of the last decade. Executives trying to launch semiconductor or OEM system companies in the new millennium automatically faced prejudice from venture backers, who favored software companies following the telecom crash of 2001. Offering a unique chip or consumer device was not good enough to win market favor. Corporate founders had to ask, can this chip be cheaper than alternative solutions? Is it easy to reverse-engineer? Will the functions become obsolete before the market develops?

Manufacturability issues

A friend of mine with decades of experience in solar design attended Solyndra’s coming-out party in 2005, and told me off the record that he was impressed by the cylindrical panels based on CIGS thin film the company had designed, “but I was aghast at all the praise and no tough questions at the rollout. The first thing many of us outside the media were thinking was, can you really build these? And where is the protection from having your cost of production undercut? I didn’t hear anyone in the press or investment banks voice these concerns.”