A Thin Film Solar Installation Revisited
A season has passed since we covered the installation of Magco Inc.’s new thin film solar panel installation. The previous article generated a lot of interest and questions, so it’s time to get back on the roof and report on some real-world figures.
If you need a refresher on thin film solar technology, check out two of our previous posts. The important thing to remember here is that thin film solar is lighter than silicon panels, and uses different wavelengths of light.
In May of 2008, Magco Inc. installed 27kW of Unisolar’s triple-junction laminate panels. That’s 4,900 sq ft of thin film goodness. The building contains a warehouse and offices, and it has a metal roof. The solar panels were literally glued to the roof.
Total installation cost: US$215,000 (including inverter and hiring a master electrician)
Energy produced each month provides about 1/3 of Magco’s total needs. That may not seem like a lot, but recall that this includes a warehouse with associated heating/cooling, machinery and equipment. Magco anticipates producing about US$9,000 each year from the panels. So let’s do the math:
If Magco earns $9,000/year, they’ll regain their investment in almost 24 years.
That’s about US$43.88 per square foot on this installation.
If you only calculate the installation cost and optimal energy production, the price per watt comes out around $8/watt.
Keep in mind that these are pre-rebate figures. State and national renewable energy programs can go a long way towards reducing these costs. Maryland does have a solar energy rebate program. More on that as the information becomes available.
Solar panels typically come with a twenty-year warranty, and they’re built to last; the wires connecting the panels will probably wear out first. These panels actually protect the roof because they’re glued right on top of it, extending its lifetime and potentially reducing maintenance costs. The panels themselves do not require any maintenance. Also, because thin film is light weight, the building did not require any extra structural support. For a large building, that’s an important feature.
Want to keep track of Magco’s solar panels? Click here for daily and monthly information and trends. As of this writing (Sept. 2008), the panels have produced 15,191kW of renewable energy.
Readers also asked important questions. Do rooftop heat islands affect performance? What about performance on sunny vs. cloudy days? Both answers relate to the qualities of thin film technology. Thin film solar panels perform well in high temperatures because of the chemicals involved. Most solar technologies (except of course solar thermal) lose efficiency at high temperatures, but the changes in performance should not be significant, unless you stick your panels in an oven.
As for sunny vs. cloudy days, the wavelengths of light that thin film solar absorbs are also the wavelengths that pass through clouds. Typically thin film solar performs better in cloudy conditions than silicon solar, but silicon panels are more efficient in direct sunlight. That’s currently the trade-off between these technologies, and it’s important to consider your local climate when choosing solar panels. Germany, for example, gets a lot of cloud cover, but that hasn’t stopped them from leading the charge to deploy solar technology.
We’ll keep our eye on Magco’s roof and report back if there are any new developments. I should also mention that Magco has other green technologies in its commercial repertoire. The same building boasts a small green roof, solar tubes, and rain barrels - oh my! Magco installed these products on their building to “go green” and better understand the real-world commercial applications of these technologies.
Have any more comments or questions? Leave a message below.
Many thanks to Patrick Bollinger and Magco Inc. for their valuable time and information!
Magco Roof photo courtesy of Magco Inc., Silicon close-up via markus941 on Flickr Creative Commons.
Related Posts:
so for middle east. with lot of direct sunlight BUT lot of heat as well. Will the heat benefit in thin films offset direct sunlight benefit of Silicon solar
MySchizoBody: I’d say that silicon solar would probably be best in a really sunny place, like the middle east. The reason is simply that the negative effects of heat are really minor. It doesn’t knock much off your total efficiency.
But naturally I’m not an expert. If you’re still concerned, ask a solar representative in your area. He’ll probably be really confused because it’s really not a big deal.
The only solar technology I’ve heard of that uses cooling systems is “extreme solar photovoltaics”, which basically put a silicon solar panel under a magnifying glass. The temperature under there is comparable to the surface of the sun, and can easily melt the panel. That’s the kind of heat you need to worry about!
How can Magco expect to produce US$9,000 worth of electricity each year from the 27KW worth of panels?
Based on the real-time data, the system has been generating on average 132KWH a day since activated in early June - this is the peak-producing summer period - so for the year, the most Magco can expect to generate is 50,000 KWHs (and this will come down every year, as the thin film they are using degrades more than 1% a year, based on NREL data).
Magco must be paying about 12c per Kwh (see http://www.eia.doe.gov/cneaf/electricity/epm/table5_6_a.html )
Thus, Magco could optimistically generate 50000 * 12c = US$6,000
Not $9,000!
Truth:
You’ve got a good point there, especially about the inevitable aging of the panels. I received that figure directly from Magco’s projections. Since the system is not yet a year old, and they’re still learning about solar technologies in real-world applications, I suspect their actual results will be closer to your projections in the long run.
As for an efficiency degradation of 1%/year, I’m pretty sure that Unisolar’s data sheets from their tests and lab results suggested a less dramatic decrease in efficiency. If I recall correctly (and I haven’t studied those sheets for a month or two), they experience degradation of .05% - .08%/ year. Please feel free to correct me if I’m wrong on that.
You can take the manufacturer’s word as you like it; I know plenty of internet-goers out there would be cynical about it. But I’m inclined to trust them within a reasonable margin. The last thing a solar manufacturer needs is litigation for false advertising.
Unfortunately, Unisolar is indeed misleading their customers about the degradation of the panels.
For example, Unisolar is committing research fraud in its Technical Report (http://www.uni-solar.com/uploadedFiles/AA53606-02Technical%20Report120706small.pdf), specifically, on page 8. The chart there (showing 0.74% degradation) comes from the October 2003 NREL (www.nrel.gov) study of dual-junction panels (a product that Unisolar stopped manufacturing years ago), not the triple-junction laminates Unisolar has been selling since 1997.
NREL has been studying the long-term degradation of Unisolar’s triple-junction laminates since 1998. For the description of the study and a sample interim report, see pages 43-48 in the above referenced Technical Report. The NREL study shows accelerating and excessive degradation, particularly after the 8th year. Average annual linear degradation went from 0.99% after 6 years to 1.14% after 7 years to 1.22% after 8 years and 2 months (that latest data point has not been disclosed by Unisolar, but can be found in a NREL presentation at http://www1.eere.energy.gov/solar/review_meeting/pdfs/exp_5_osterwald_nrel.pdf on page 9). Unisolar has declined to disclose the most recent (June 2008) results.
Unisolar’s warranty implies less than 1% annual linear degradation (as it is a 20-year warranty for 80% rated power), so the NREL study indicates a likely violation of Unisolar’s warranty.
i applaud Magco’s efforts, renewable engry is the future and our survival, you should all so mention the amount of CO2 and biproducts not being released into the air we breath. It is not only a renewable source of energy it helps renew my faith in a better future.
To John Benjes:
Unfortunately, careful cost-benefit analysis has indicated that the current solar technology at the current prices actually wastes productive resources, even after considering greenhouse gas reductions:
http://www.newswise.com/articles/view/537915/
It should be noted that it takes between one and three years for Unisolar laminates to recoup the energy used to produce them. Steel is about 20% of the costs to produce Unisolar laminates, and the steel industry now accounts for four percent of global man-made greenhouse gas emissions (double what cars and light trucks contribute).
@ John Benjes: glad to renew your faith in humanity. = )
@ TruthSeeker: I applaud your research and investigation into the issue, but I’d like to pick apart your conclusion based on Mr. Borenstein’s paper. According to the article, Mr. Borenstein looked at PV solar (not thin-film), and his conclusion was that the reason PV solar’s value was comparatively low was due to outdated infrastructure which is not currently designed to make best use of decentralized power generation.
This is not and has not been news for quite a while. I’ve written on the topic by covering a debate hosted by “The Economist” in which both sides agreed that our current outdated infrastructure is probably the greatest hindrance to renewable energy dispersal today.
Mr. Borenstein concludes that if CO2 emissions are included in calculations of value, PV panels are still better than coal-fired plants. As such, they still make a better investment for consumers, if even on a small scale.
Nevertheless, there has been increasing investment in utility-scale solar plants, predominantly in solar-thermal technologies, which require simple materials like steel, compared to PV solar cells. Large scale solar facilities increase the value of solar energy production and provide a more centralized power grid model, which our current infrastructure can best take advantage of.
I will let Mr. Borenstein speak for himself (see the original paper, the critique, and his response at
http://faculty.haas.berkeley.edu/borenste/PVwork.html ). If you read carefully, you will see that he is talking exactly about rooftop installations like Magco’s, not about large ground-mounted PV farms.
You have misread the paragraph in the article addressing CO2 emissions and coal-fired plants (see the details on page 25 of his orginal paper). Mr. Borenstein actually says that since the permit prices per ton of greenhouse gases are $20-$100, PV’s costs still DO NOT justify the benefits vs coal-fired plants - they might only if and when the greenhouse prices ever get to $300.
Thin-film (and Unisolar’s triple-junction a-Si, in particular) is just one of the many PV solar variants - mono-crystalline (high efficiency), poly-crystalline (the bulk of the market today), thin-film CdTe (the lowest cost), thin-film a-Si (the easiest to produce from scratch). It all comes down to all-in system cost per KWHs generated over the warranted life of the system. It should be noted that Unisolar’s laminates lag the rest of the industry in terms of that metrics. Degradation over 1% a year (based on the NREL data) vs typical 0.5% for crystalline, combined with the short warranted life of 20 years vs 25 years for crystalline, means that Unisolar laminates glued to a flat roof generate about 20% less KWHs over the warranted life of the system than tilted crystalline and about 30% less than tilted thin film (either CdTe or a-Si).
At about $8 per Watt, the Magco system has little chance of paying off, as the payback period far exceeds the warranty (and we haven’t even accounted for the need of inverter replacement, as inverters are typically warranted for 10 years only). That’s what Mr. Borenstein was essentially talking about. Those $215,000 might have been used for more productive purposes.
TruthSeeker’s sure doing the homework here. Wow.
On the other side, it’s going to take some early adpoters to move this technology along, and probably will take some people losing some money to learn along the way.
Out of curiosity TruthSeeker, (and this isn’t an attack at all, just curious) what would you suggest would be a better alternative, since you do know a lot about the topic.
Thanks!
Jonathan
http://www.GreenJoyment.com