Some interesting things have recently happened in the world of solar power: Evergreen
have gone bankrupt, panel cost has gone sub $1.00/watt, and China has vastly increased production capacities.
Before getting to the much-publicized Solyndra, let's see what happened to Evergreen. Evergreen built a hugely expensive facility
in the Boston area. Certain parts of Evergreen's assets have already been auctioned off, including 7.8 megawatts of 200W to 215W panels
which were purchased by SunElec, a wholesaler in Miami. These were sold off in a couple of days at 78 cents/watt, minimum quantity one pallet.
What went wrong with Evergreen? They received tax breaks
from the state of MA to set up a facility there. Unfortunately, economies of scale couldn't compete with vertically integrated solar firms in China. Their string ribbon
panels maxed out at a wattage density of around 130 to 135W/square meter, while current generation budget panels from China are pushing 155 to 165W/square meter.
In the past year, solar panel costs from China have dropped from $1.65/watt to sub $1.00/watt
. This has driven prices down worldwide. It's now possible to buy low quantity (quantity, one!) solar panels made by Sovello
for $1.04/watt from US distributors. The relative simplicity of assembling a solar panel of 60 or 72 156mm cells has lent itself to highly automated production in China. Basically, take a plastic solar module backsheet
, place a series of aligned and tabbed
solar cells on it, laminate with EVA, add aluminum frame and rear junction box, package the finished product.
For the electronics manufacturing industry in China this has not been difficult to scale up to very large size plants. Making a solar panel is actually quite a bit less complicated than making a basic multi layer $80 ATX motherboard for a desktop PC.
How quickly has China scaled up? Canadian Solar recently supplied the modules for the world's largest solar plant
in southern Germany. The cost drops on solar PV mean that the Blythe Solar Power Project
, which may cost as much as $6 billion, has recently changed plans from using concentrated (mirror-reflected, steam turbine) concentrated solar to PV. First Solar
has put out a few press releases about supplying up to 1.0 gigawatts of CdTE thin-film modules for the Blythe project.
And now for Solyndra. I was recently asked:
"Tell us about why Solyndra was a terrible idea both technology and economically. I have my thoughts on the matter but would love more input from a other person in the field".
Politics aside, here's my technical opinion (as a person that works with solar) why Solyndra in general was a bad idea:
1) The weird cylindrical tube manufacturing process was very expensive both in capital equipment costs, time, and labor.
2) The relatively low power density of the individual assembled modules meant that it was only economical for very large roofs. It had a density more comparable to amorphous Si panels on glass, not 155W/square meter polycrystalline Si 60 or 72 cell modules.
3) The power rating of each module was only with a white painted roof. If you look at photos of the larger examples of Solyndra installations the roof was always painted reflective white, and this was required for any large installation. This added to installation costs and also meant that any installed solyndra system would slowly start generating less power every year as the roof got dirty. There was no practical way to pressure wash a white roof once it was covered in wired-together modules.
4) Its lunch was eaten by $1.20 to $1.50/watt poly and monocrystalline Si modules from China that anybody could buy, while Solyndra continued to only sell its modules through specially approved dealers and at mounted costs of much higher than $2.00/watt for just the modules and feet.
5) It was only suitable to one type of roof. You can mount a 60 cell aluminum framed module in almost any way - on your house, on angle mounts on top of a home depot, on the bimini top of a sailboat, vertically on the wall on the side of a telecom equipment shelter, etc.
Basically, Solyndra created a completely proprietary product that had a terribly expensive manufacturing process, and built a huge factory
to scale up their technology at exactly the same time
that various Chinese competitors were scaling up the mass production of 156mm polycrystalline and monocrystalline cells. Digitimes, a Taiwanese semiconductor industry trade journal
reports that polycrystalline 156mm cells are now sub $0.55/watt, and monocrystalline cells are approaching the $0.60/watt figure.
Three US firms are working on thin film technologies which have the possibility of approaching the magical $1.00/watt figure and even going below it.
, Global Solar
have built pilot lines which produce flexible CIS/CIGS solar modules that can be glued directly onto flat roofs. These come in rolls, with a single Global Solar module measuring 5.7 x 0.5 meters and with a wattage density of about 106W/square meter. Solopower has received a $197 million loan guarantee
to build a fab in Oregon to produce their modules. They're not shipping in commercially viable quantities yet, but all three of these companies (which make a very similar product) are going to face stiff competition from the abovementioned $1.00/watt 300W solar modules from China.
On the high end: Sunpower has produced a 327W panel
now commercially available which has a density of 200.5W/square meter, far exceeding anything available from China. This is produced at a fab in Malaysia, in a joint venture with AU Optronics
, a major manufacturer of flat panel LCD displays. If the race to the bottom on price isn't a viable business plan, some firms have decided to compete with better technology. The Sunpower modules are particularly interesting as the positive and negative contacts for each cell are entirely on the back sides of the cells, eliminating the vertical lines seen on common solar cells. Every few percent of added efficiency is helpful.