Recent events in Solar Power

February 27, 2010 12:24 PM

Solar power continues to become more economical with every passing month. The price of standard monocrystalline/polycrystalline 210W to 300W modules has halved, from above $3/watt to below $1.70/watt in the past 12 months. The Canadian province of Ontario has become the first location in North America to offer an aggressive feed-in tariff comparable to the feed-in tariffs in Germany and Spain, paying above 80 cents/kWh for power generated by small rooftop photovoltaic installations. The US federal government is also offering massive tax breaks for solar installations as part of the 2009 ARRA act.

The new Ontario feed-in tariff (PDF) pays between 70 and 81 Canadian cents per kWh for rooftop power generated by photovoltaics. Payments are lower for power generated by ground mounted installations, and the payments decrease as the system size grows larger than 250kW. There is also a requirement that "Class 1" and "Class 2" high quality farm land must not be used for ground based PV installations. The Ontario FIT's domestic minimum 50% content requirement means that installers will need to purchase Ontario-manufactured solar modules, source their cables and interconnects from Ontario sources, and perform the installation entirely with Ontario-based labor. Companies such as Canadian Solar have announced plans to build 200MW+ of module assembly capacity in Ontario to meet this domestic-content requirement. In this arrangement, cells will be imported from the factory in China and assembled into whole modules at the plant in Ontario.

Chinese based vertically integrated solar cell+module manufacturers such as Trina Solar, JA Solar and Suntech and a vast oversupply in the global market are driving costs even lower, past the $1.70 USD/watt mark. Dozens of smaller firms in China are also sourcing 125mm or 156mm cells from 3rd parties to assemble modules in competition with the larger companies such as Suntech.

Big players in the US thin-film manufacturing market, First Solar and NanoSolar both both claim to have achieved and passed the magical $1/watt manufacturing mark. With 2009 revenues of $2 billion USD, First Solar is becoming the 500 pound gorilla in the thin-film market.

PV modules from reputable manufacturers are mostly all warranted for 80% of original power levels after 25 years of service life. Even without government subsidies, in the right climate a grid-tied residential solar system can now have a lifetime kWh cost below 15 US cents/kWh. This $0.15c figure also assumes that the system will have zero scrap or resale value after 20 or 25 years of service...

posted by thewalrus (50 comments total)

Wow!! Thanks for this massively informative FPP, and good news to boot.
posted by tybeet at 12:43 PM on February 27, 2010 [3 favorites]

Great post, thanks. The feed-in tariff is certainly encourages adoption of solar technologies, but as is the nature of government programs, it risks not being able to respond quickly enough to changing market conditions to safeguard the Ontario ratepayer. Eighty cents per kilowatt hour is extremely high. By way of comparison, wind power plants receive 13.5 cents per kilowatt hour, and nuclear power operators receive five cents per kilowatt hour. In fairness, p 14 of the pdf linked on the page in the "province of Ontario" link shows that the price for solar varies; larger ground installations will receive 44 cents/kWh, still about nine times more than a nuclear plant would.

The real rationale for this only exists if you consider that there is or ought to be a high price put on carbon, because solar power will, in theory, replace gas and coal power that operates at times of peak demand (coal and its associated pollutants will be phased out by 2014 in Ontario, if all goes according to plan). What it does do is spur innovation in the solar field, because the cheaper you produce your power, the more money you make. Still, if thin-film solar comes online in a big way and traditional photovoltaics become obsolete, I really hope that the government doesn't leave the tariff where it is, because those costs are so high that even if solar only makes up a couple of percent of the grid, power rates will go up noticeably. Germany has been having this debate.

The other issue with widespread solar power is that the grid is not designed to cope with many distributed power inputs; it's designed to take power from a few big stations and distribute it out from there. If thousands of people start putting solar panels on their roofs, and companies start building solar panel installations all over the place, the grid in its current form is not going to be able to deal with it. And upgrading the grid - and getting approvals to build new power lines all over the place - is not going to be cheap or easy.

Still, cheap solar is better than expensive solar. Carry on.
posted by Dasein at 12:48 PM on February 27, 2010

I support pollution taxes for dirty power, but feed-in tariffs for green technologies is a horrible idea. Last year at the depth of the recession Ontario's power demand dropped substantially. However the government had contractual obligations to accept wind and solar generated power at three to ten times the commercial rate. The effect was to force the big baseload power stations (hydro and nuclear) to dump water over the sluice gates or into the condensors - a process which costs the utilities money and shortens the lives of these plants. In fact, the cost of power went *negative* briefly, as the Ontario Power Authority tried to drum up enough demand to take the uneconomic wind power. In a sane world, the turbines would have feathered their blades during periods of low demand, but that wouldn't have satisfied the investors in these sweetheart deals.
posted by Popular Ethics at 12:51 PM on February 27, 2010 [4 favorites]

If an extremely high FIT drives up the average power bill too much, and solar panels are extremely cheap, people will just buy the solar system and disconnect from the grid... Spend $20,000 one-time cost to have 25 years of 15 cents/kWh electricity.

I don't think many distributed power inputs will be a big problem as long as -most- of the power generated in a house is used in the same house. If I have a solar system that will generate about 950kWh a month and my house uses an average of 800 to 850 kWh, the local power utility doesn't have much to worry about, unless solar rooftop installations reach 100% market penetration in my neighborhood. If electric cars like the Volt catch on houses will probably remain net consumers, not producers of electricity (a PV system that can produce 1500kWh a month from October to March at 45 degrees north latitude is quite expensive and needs a large rooftop area, barring advancements beyond 18% efficiency for monocrystalline cells).
posted by thewalrus at 12:58 PM on February 27, 2010

Dasein : The other issue with widespread solar power is that the grid is not designed to cope with many distributed power inputs

With solar in most parts of the US, that doesn't present much of a problem - You really need to try to exceed your normal daily consumption. Home solar mostly just reduces how much you draw from the grid, without feeding much (if any) back.

And, conveniently, the places where solar does work best tend to need to run their ACs the hardest when solar produces the most (middle of the day in the summer).

I do agree with you in the abstract, though - We have an entirely too fragile power grid. Reducing the load on large plants by encouraging massively distributed baseline production (solar and small-scale wind) seems like a no-brainer, but any net-positive production causes more trouble than it saves at this point. We need that fixed ASAP so we can start actually using some of the wide variety of alternative power sources starting to hit widespread availability.


thewalrus : If an extremely high FIT drives up the average power bill too much, and solar panels are extremely cheap, people will just buy the solar system and disconnect from the grid... Spend $20,000 one-time cost to have 25 years of 15 cents/kWh electricity.

I've personally looked into the options for doing exactly that, even if it costs more than buying from the grid - And the options look pretty bleak. First, even with solar+wind (which unless you have some uncommon and almost limitless natural resource available to you, gives the best realistic balance of day/night production), you need some pretty hefty batteries. Second, even where both will work, they don't always work all that well - Expect to start buying a lot more fossil fuels to cover your home energy demands (ie, propane powered dryer and refridgerator). And finally, a scary number of places won't let you use alternative energy sources, whether for purely cosmetic reasons ("no wind turbines in my backyard, ya damned commies!") or pseudo-safety (yes, you need to pay a real electrician to hook you up, even if you go totally off-grid, and "no applicable codes to enforce" in many places means "you can't do this").

So for now, the most realistic option looks like going with a grid-tie inverter, which means the utilities will just start charging more for their "service" than for your actual consumption.
posted by pla at 1:11 PM on February 27, 2010

If an extremely high FIT drives up the average power bill too much, and solar panels are extremely cheap, people will just buy the solar system and disconnect from the grid

Well, no, because the other problem with solar power is that it's intermittent. You're limited by the fact that at night and on a cloudy day, you're getting nothing or not much. So you need batteries, which are really expensive for storing a decent amount of power, and which wear out. This the other, hidden, cost of solar and wind power - the amount of money you pay to have all those gas plants standing around idle, just waiting to be fired up. If the plants are built privately, then you generally have to guarantee the operators a certain amount of revenue to make their investment worthwhile, irrespective of whether you actually end up demanding their power. Some of that is a matter of how you structure your electricity market; but until we come up with much, much cheaper storage solutions for electricity, solar and wind will remain a very expensive way to offset some amount of gas generation, not a replacement for having gas plants around in the first place. (Not that people aren't working on it.)
posted by Dasein at 1:16 PM on February 27, 2010 [1 favorite]

pia: Yes, it costs more than buying power from the grid. The advanced charge controllers ($600 for an Outback 80A) and big bank of 220aH 6V batteries wired up to make a 48V string do add a lot of extra cost. Over 16 years it works out to something like 25 US cents/kWh, and you'll need to replace the batteries after each 8 years of service, assuming the use of very high quality batteries. That's for a system that will produce enough power to meet the needs of an average home without electric heating of any sort (hot water heaters, home heat, stove). You would still need to remain connected to the municipal gas supply line.

The solar systems I have built professionally for off-grid telecommunications tower projects are mostly alternatives to generators. Using a 20kW diesel generator for some projects has a minimum cost of about 45 to 50 US cents/kWh, and it'll fail mechanically after 5 to 8 years of constant duty, even with regular skilled maintenance. The solar systems in the 10kW range come in under 30cents/kWh...

My home is still 100% grid powered. :-(
posted by thewalrus at 1:17 PM on February 27, 2010 [1 favorite]

I want to add that here in California, solar makes a lot of economical sense, especially for heavy electricity users. Through the California Solar Initiative, there are significant incentives, although they decline as we get closer to our statewide solar goal. Right now, it's $1.10-$1.90/watt for residential, depending on your utility. For grid-tied systems, the utilities will pay you the retail rate of electricity if you are a net producer.
posted by kookaburra at 1:42 PM on February 27, 2010

Crystalline silicon panels prices have started firming in the last six weeks and in some cases have increased (for Q4 2010 to Q1 2011 delivery). Talking to a major Chinese panel manufacturer this week, they claimed that announced PV projects account for most of current inventory plus this year's estimated production. Who knows. We're seeing prices in the range of $1.60 to $1.80/Wp (e.g. W (DC)) right now.

On the utility-scale side, financing is the biggest hurdle. Banks are still very risk adverse and the cost of financing is high enough that traditional tax equity financing structures don't make sense, even with the cash grant. The major PV projects that have been announced during the last year are not, in fact, under construction and most of these projects we think assumed the availability of the DOE loan guarantee. And maybe they're right; Brightsource announced that it had received $1.3 billion this week from the DOE (a loan guarantee) for its Ivanpah project (which is mired in a permitting nightmare because of 25 desert turtles). But it's not clear, despite Brightsource's announcement, that all of these projects will ultimately get the loan guarantee and I'm increasingly convinced that many of these projects are not economical without it.

On the grid: I know just enough about transmission to be dangerous, but integrating wind or solar PV poses real challenges to grid operators. One area of concern on the utility-scale solar projects is the impact of clouds: output can drop suddenly and it causes real instability on the transmission system. There are similar issues with wind, including an incident in early 2009 where the sudden, unexpected drop in wind from the Columbia River Gorge almost brought BPA's system down.

I can provide links from work computer later if folks are interested.
posted by jchilib at 1:58 PM on February 27, 2010 [6 favorites]

I'm quite pleased that Ontario has made moves in support of alternative energy, especially as the conservative Canadian federal government has done little on the industry brief besides bragging about oilsands and our cautious (pronounced "solvent") banks.

Besides the Green Energy Act , the Ontario government has inked a $7B deal with Samsung that will see 2,500 megawatts of solar and wind developments and construction of four manufacturing plants between 2013 and 2015, and allegedly create 16,000 jobs. This new deal has drawn criticism, but it has also attracted new investment in Ontario from players in the alternative-energy field.

Regarding feed-in, I believe the current grid is currently more than sufficient to handle the feed-in from residential and small-player PV and wind systems, since their output will always be only a small fraction of the energy going into the grid. Bigger installations, like wind and solar farms will of course be connected to the grid as net suppliers.

The details might need tuning, but this is exactly the right thing for Ontario to be investing in.
posted by Artful Codger at 2:25 PM on February 27, 2010

Toughstuff do something I like, if you buy a charging kit for $30, they'll outfit an entrepreneur in Mozambique with an identical kit
posted by infini at 2:26 PM on February 27, 2010 [1 favorite]

Feed in tariffs are also regressive. They are funded by a charge on everyone's electricity bill, but the substantial benefits they provide tend to accrue to a few wealthy people and to companies. The international energy agency has also pointed out that FiTs are a very expensive way of reducing emissions. (I'd link to the reports but am posting from phone, can do so later if people are interested).

It's not widely recognised that germany's primary motivation for a FiT was to help provide energy independence from russia.
posted by girlgenius at 3:10 PM on February 27, 2010 [2 favorites]

Lower prices for solar is always good news. Especially here; I live in southern Arizona. We get 350 days of sunlight a year, and many of those days the sun is insanely intense. I've been frustrated at the pace of solar development (or lack thereof) over the past few decades and also the cost of panels. We have gigawatts of free energy available from the sky almost every day, but the energy emphasis has been on oil, gas, and coal, so we miss out on this opportunity. Instead, I have a power plant a mile away from me, and sometimes we're treated to the lovely scent of what smells like asphalt from there.

As soon as we can afford 'em, we wanna get 'em. I've got 1500 sqft of flat rooftop with nothing blocking the southern sky.
posted by azpenguin at 4:06 PM on February 27, 2010

I love solar power--so cheap. I dream about it every night.

Thing is, I can't exactly fork out $25K right now (and then another $5K on).

That's ridiculous anyways and not fair. I should not have to pay so much up front. It's just energy--powers my home so I can enjoy a higher standard of living. Since when did a higher standard of living get so expensive?

Hopefully the federal government will pay for it all they said they would--and my healthcare too. That's not too much to ask. I deserve it. No, I'm entitled to it.

Energy and health--one should not have to pay for things like this.
posted by stevenstevo at 4:38 PM on February 27, 2010

Do you drive on the public roads, eat food products that are regulated by the FDA, fly in airplanes regulated by the FAA or have a municipal water supply connection to your house?
posted by thewalrus at 4:40 PM on February 27, 2010 [2 favorites]

FYI, there are alternatives to silicon photovoltaics ...

Gallium arsenide (GaAs) photovoltaics with optical concentrators.
posted by ZenMasterThis at 4:48 PM on February 27, 2010

I was recently asking about the Ontario FIT on behalf of the condo building where I live. The cell prices are getting better and better, but it's still an expensive proposition (we were looking at a ~50kw install, so the ballpark quote was $400k).

As has been mentioned above, this is going to be a big win for those that can afford it. Unfortunately we have more pressing issues to deal with.
posted by lowlife at 5:01 PM on February 27, 2010

In US dollars, 50kW is about $97,500 of panels alone, at $1.95 USD/watt (to buy Ontario-made panels, to qualify for the 50% Ontario content rule). That's for very generic panels that each use 60 or 72 monocrystalline 156mm cells.

That 400k quote seems really high even including labor, mounting, cabling, junction boxes, grid-tie inverters. If you qualify for the 70 cent FIT rate the quote should be low enough that the system will pay for itself in 12 to 13 years, while the FIT contract is for 70c/kWh for 20 years. Profit!
posted by thewalrus at 6:14 PM on February 27, 2010

thewalrus: I recognize the opportunity for sure, we were hearing an all-in price of around $8/watt (including all the studies, additional hardware, planning permission, blah blah). I agree that, long term, this would be a great thing for the building. Short term, we can't just find 400k and I don't know if we can go around cap-in-hand to various commercial financing orgs.

I will continue to push though ;)
posted by lowlife at 6:39 PM on February 27, 2010

We have gigawatts of free energy available from the sky almost every day

It's a mistake to think like this. It's not really free. It costs money to harness it and convert it to electrical power. It also costs money to "gather" it, because it's such a widely dispersed resource. If you wanted to think like that, you might as well call all energy sources free. After all, fuel is just sitting in the ground, waiting to be gathered up and used, in the same way that light is just "sitting" in the sky.

A better (and less frustrating) way to think is in terms of what is realistically achievable and at what cost. Solar PV might never be cheap enough, and I think it's unhealthy the way some people yearn for it like it's the only possibility and guaranteed to solve everything eventually. Solar PV may or may not turn out to be the best solution to the energy problem. We should keep our eyes and minds open to other possibilities. What if ITER is a stunning success and fusion solves everything? Or wind? Or tidal power? Or Solar thermal? Osmotic power? Geothermal? Clean coal (yes, it does actually exist, regardless of any debate over its potential)? Nuclear (fission)? Biofuel? All of the above?

I'm keeping an eye on all of them.
posted by Xezlec at 7:05 PM on February 27, 2010 [2 favorites]

"It's not widely recognised that germany's primary motivation for a FiT was to help provide energy independence from russia."

In the US we could sure use ways to keep $600B from sailing off into the distance each year.

Another way of looking at it: at $70/kWh it's not hard to pay for a $100K/100kW array in 2 years. Manufacturer guarantees for 25 years. On- or off-grid (your neighbors' EV's?), that power won't go wasted. Compare to (long-term) cost of a 1GW nuclear plant.
posted by Twang at 7:14 PM on February 27, 2010 [1 favorite]

Oops that's $0.70/kWh.

(WHEN will MF EVER allow editing posts for 5 or 10 minutes?)
posted by Twang at 7:16 PM on February 27, 2010

Companies such as Canadian Solar have announced plans to build 200MW+ of module assembly capacity in Ontario to meet this domestic-content requirement.

Heh. Most people probably don't realize this, but "Canadian Solar" is actually a Chinese company. They have offices in Canada, but all of their work is actually done in China. In fact, I think most of their sales are done in China as well. From the bit on yahoo finance:

In addition, it implements solar power development projects, primarily in conjunction with government organizations to provide solar power generation in rural areas of the People's Republic of China.
...
t has a strategic cooperation agreement with Guodian Power Development Co. Ltd. to design, invest, build, operate, and maintain solar power plants in China.

(they also sell various modules)

---

That said, I've always wondered why people don't use concentrators more. The opel solar thing that ZenMasterThis linked too always made sense to me. If the panels are so expensive, why not use cheap mirrors to increase the amount of light they take in?
posted by delmoi at 7:25 PM on February 27, 2010

If the panels are so expensive, why not use cheap mirrors to increase the amount of light they take in?

I've wondered this too, but my best guess would be that the greater insolation increases the power output less than you would think, and also decreases the lifetime of the panel, thus increasing your effective amortized capital costs. I base the former hypothesis on the fact that, in general, semiconductor device performance drops very rapidly with increasing temperature, and the latter one on the fact that microprocessors and light bulbs and stuff die more quickly when you run them hot.
posted by Xezlec at 7:37 PM on February 27, 2010

Hot damn. My home is flat-roofed. If I can turn it into income, that's just grand.
posted by five fresh fish at 8:05 PM on February 27, 2010

I'm not going to be happy until we build massive solar arrays in space and then beam the power down to earth. That such an array can double as a doomsday weapon is an added bonus.
posted by Justinian at 8:07 PM on February 27, 2010

Twang: I think the problem is that $100k doesn't buy 100kW of array, $100k USD buys about 50kW of panels these days... Then you need to double the cost to account for shipping, labor, mounts (fixed), connectors, nuts/bolts, cables, junction boxes, breakers/disconnects, conduit, and some beefy grid-tie inverters. The only other problem is that these really high Ontario FIT rates are only for roof systems. Great if you happen to own a 50,000 square foot warehouse with a flat roof.

Concentrators need to be mounted on two-axis tracking systems that can exactly follow the sun across the sky. Compared to cheap fixed mount angled panels on a roof ( such as this type of setup the tracker and its foundations drive the cost up a lot. IMHO concentrated PV, such as a combination of fresnel lenses and GaAs triple junction cells made more sense when monocrystalline panels were $3.50/watt.
posted by thewalrus at 8:11 PM on February 27, 2010

> The other issue with widespread solar power is that the grid is not designed to cope with many distributed power inputs

The Ontario grid isn't, but it's currently not in much of a state to be designed to do anything. The planned upgrades that were supposed to happen when Ontario Hydro split? Evaporated. The only upgrades that are really likely to happen is the Bruce-Milton line, and that's basically to keep renewable energy's mucky paws off Bruce Power's the province's 500kV line down through Huron.

> The effect was to force the big baseload power stations (hydro and nuclear) to dump water over the sluice gates or into the condensors

Ontario's currently got just over 1200MW of installed wind capacity. The provincial demand seldom (if ever) drops below 10000MW, so even if all the wind were running flat out, wind would provide 12% of total demand. Don't you think problem might more likely be the province's huge nuclear plants that can't respond to demand changes? You can feather wind turbines, and under new market regs you soon will have to, but it's motes and beams.

> the amount of money you pay to have all those gas plants standing around idle

That's at the owner's risk, not the province's. Most gas peakers are designed and costed to run at around 10% CF.

> Feed in tariffs are also regressive. They are funded by a charge on everyone's electricity bill

Yes, but we also have a Debt Retirement Charge on our bill in Ontario. The former Ontario Hydro had biblical cost over-runs on its projects, and we're still paying it. I'm sure the debt retirement, at around $5/quarter for me, is far more than any FIT would cost me.

Renewable energy brought at least one job to Ontario ... mine. When I came here nearly 8 years ago, I wasn't able to do my chosen profession, and was making do in IT.

> a ~50kw install, so the ballpark quote was $400k

We need to talk. My farmers on the wind farms are chasing off solar dealers with shotguns and rocksalt there are so many of them. Guide price for a 10kW installation on the farm is $80-100K. Upper range is for trackers, which I have my doubts about their survivability by windy Lake Huron.
posted by scruss at 8:31 PM on February 27, 2010

Jcheleb : please post links.

This is a great discussion, people.
posted by stratastar at 8:56 PM on February 27, 2010

These guys seem to have the best prices I've seen so far. Worth the drive to Florida from anywhere on the Eastern seaboard.
posted by CynicalKnight at 10:24 PM on February 27, 2010

Sun in Miami seems to have the lowest direct sale to consumer prices in the US right now. The only way to get mono/polycrystalline modules cheaper is to buy about 30 cubic meters of them in a 20' cargo container from China.

For those who care the Sun Electronics ($1.74/watt) house brand panels are Evergreen "B" grade panels. Not "B" grade because there's anything wrong wattage/electrically with them but because the color density/shading of the polycrystalline cells doesn't match perfectly, so they're considered blemished... Unless you are supremely picky you won't care about it once it's installed on your roof.

But anyone thinking of buying the $1/watt Kaneka GSA60 panels should consider the total mounting cost - an 8kW system is one hundred thirty four panels, you'll want a large roof.
posted by thewalrus at 11:18 PM on February 27, 2010

Xezlec, I see your point about the cost of harnessing solar. And that's been one of my beefs with the whole thing - up until recently, the cost hasn't really changed much. Over the last several years, R&D into solar has picked up quite a bit. But why now and not before? Bringing the cost down would be a game changer as far as energy goes.

I don't see solar or any one thing as the solution to energy. I see a multitude of sources as being our energy future. As it is, we have plenty of solar energy available here. There have been stretches in the summer where there's been hardly a cloud in the sky for weeks. During peak demand (summer days), solar would be providing peak power. Since I run a swamp cooler, I can spin the meter backwards with only a few kW on the roof. It's not hard to see why I'm interested in solar.

Last, yes, there's costs involved with solar. But it's pretty much all one-time. There are no mechanisms to oil. No moving parts. No need to use up precious water to drive turbines. No emissions. No waste material (burnt coal, spent rods) to dispose of. No transportation of fuel. No mining. No drilling. No pollution. No nuclear safeguards needed. No resources that we need to expend military might on protecting. No habitat altered for any animals. Plus the bonus that you're using the same energy that heats the house up to cool the house down. Once you've paid the initial costs, that's free energy. Free from a lot of problems. And bringing down those initial costs are a goal worth pursuing. Bringing solar online in an area that needs extra energy because of the sun would free up a lot of power generation that can be used elsewhere.
posted by azpenguin at 11:26 PM on February 27, 2010 [1 favorite]

the substantial benefits they provide tend to accrue to a few wealthy people and to companies

That's an interesting idea. Where do the benefits of fossil fuel electric plants accrue? And do the benefits of reducing carbon only accrue to a few wealthy people and to companies?
posted by Jimmy Havok at 12:19 AM on February 28, 2010

One of the reasons of the recent drop in PV price is the "collapse" of the Spanish market (due to a change in feed in tariffs regulation and the economic crises): the PV Spanish market absorbed half of the worldwide production (2511MW) in 2008. Add to this the irruption of cheap PV from China and you explain a bit part of the drop in prices.
posted by samelborp at 3:16 AM on February 28, 2010

Feed-in tariffs for the E.U.
posted by ersatz at 4:02 AM on February 28, 2010

Xezlec: GaAs operates efficiently at much higher temperatures than Si. So concentrators make more sense for GaAs.
posted by ZenMasterThis at 6:02 AM on February 28, 2010

a different Pov

Will developing countries that have not yet built out the power grid to much of their population completely skip the traditional power infrastructure and turn directly to distributed solar for power generation? Several analysts and executives recently have told me “yes,” and it’ll happen sooner than we think. /Via and gratuitious /via
posted by infini at 6:24 AM on February 28, 2010

irruption? eruption or interruption? as far as I know the chinese PV supply is doing the opposite of being interrupted...
posted by thewalrus at 6:42 AM on February 28, 2010

Some links:

A window into the grid issues caused by intermittent renewable energy. This is a wind example (but solar is similar). See in particular nos 13 and 14 that show sudden increase and decrease in wind production from the Columbia River Gorge. This sudden fluctuation requires the transmission operator to increase/decrease other power plants. The takeaway is that in order to integrate intermittent renewable resources and maintain grid stability, the grid operators needs non-intermittent power plants that can be quickly ramped up and down as a back-stop. Currently natural gas fired peaking plants are best capable of providing this service. And we expect more natgas plants to enter into be built just for this purpose.

On Brightsource's Ivanpah solar project, the environmental issues center around the desert tortoise (25 of them). Local opposition seems to be mounting, and possibly should be viewed within the larger context of opposition to solar development in the desert.

I don't have any links to the recent firming/increase in PV panel prices for near-term delivery because it's all quotes we've received directly. Sorry.
posted by jchilib at 7:10 AM on February 28, 2010 [1 favorite]

GaAs operates efficiently at much higher temperatures than Si. So concentrators make more sense for GaAs.

OK, thanks. Didn't know that. My comment was kind of a guess.

Over the last several years, R&D into solar has picked up quite a bit. But why now and not before?

I thought it was because of the delayed effects of Germany's (and others') solar programs funneling money into solar panel companies.

Last, yes, there's costs involved with solar. But it's pretty much all one-time. There are no mechanisms to oil. No moving parts. No need to use up precious water to drive turbines. No emissions. No waste material (burnt coal, spent rods) to dispose of. No transportation of fuel. No mining. No drilling. No pollution. No nuclear safeguards needed. No resources that we need to expend military might on protecting. No habitat altered for any animals.

In economics, costs can be translated back and forth between one-time and recurring by a formula. (Sorry if you already know all this, but for the benefit of anyone who doesn't...) A one-time expense of $100 might be roughly equivalent to a recurring cost on the order of $10 per year for eternity. This sounds very strange at first, but the reason is that there are lots of things you could do with your time and money now, and many of them pay you back forever. On average, the best of these things might earn you around 10% per year in perpetuity. So, by spending it on something else, you've lost the opportunity to earn that 10% per year. Therefore, whatever you're doing with your capital instead had better be worth losing than 10% per year you've missed out on.

Another thing is that photovoltaics don't have unlimited life. The lifetime I usually see on the web is about 25 years. So besides the 10% per year or so above, you have to add in the initial expense divided by 25, because that's how often you'll need to spend it again.

The manufacturing costs of building the panels must therefore be incurred repeatedly, every 25 years or so, and those costs may well include mechanisms to oil, moving parts, emissions, waste material, and transportation, and certainly include mining, drilling, and resources. What's more, the solar panels probably would take up space (and therefore habitats). I don't know whether we can power our civilization entirely with rooftops.

Many of the above things are captured by economics and represented in the price of solar versus other sources. That's why I was saying that looking at the price may make more sense than sitting and thinking about specific consequences and trying to say that one power source just seems better than another. The broader picture may be too complex for that kind of thinking. As unpopular as it may be in this crowd, economics is a powerful tool for tracking the costs of complicated choices (if we as a society can agree on a good cost function, that is).
posted by Xezlec at 8:25 AM on February 28, 2010

> I don't see solar or any one thing as the solution to energy. I see a multitude of sources as being our energy future.

Absolutely. There ain't but the one way.

> See in particular nos 13 and 14 that show sudden increase and decrease in wind production from the Columbia River Gorge.

It looks like the original premise is the confusion of wind for a capacity source. It's an energy source; best to think of it as negative load. BPA seems to have more than 10% of its installed capacity from wind, yet appears to have no ramp rate regulation (built into most wind farm SCADA systems) and no forecasting. They appear to be addressing these system shortcomings [PDF].
posted by scruss at 9:30 AM on February 28, 2010 [1 favorite]

On average, the best of these things might earn you around 10% per year in perpetuity. So, by spending it on something else, you've lost the opportunity to earn that 10% per year. Therefore, whatever you're doing with your capital instead had better be worth losing than 10% per year you've missed out on.

Mmm, yeah...can I get in on that 10% a year thing you've got going? In other words,you pulled that unrealistically high number out of...well, it wasn't anyplace reputable.

So besides the 10% per year or so above

And having pulled it out, now you're treating it as if it was a legitimate figure.

Your principle is correct, we have to deal with opportunity cost, but the average opportunity cost of an investment is only very rarely 10%. It's more in the range of 5% at the maximum, and even that has to be amortized against risk. If opportunity cost was 10%, you wouldn't be able to find anyone to lend you money for less, since they'd be putting their money into that sweet 10% deal.

Then you have to look at the cost of alternatives to your course of action. Not building non-petroleum energy sources isn't free, since the capacity has to come from somewhere, so you can deduct the cost of building that conventional plant from the cost of the non-petroleum plant, as well as the environmental cost (which is currently hidden pretty well) of that conventional plant.

You're right, economics is a powerful tool...but only as powerful as the accuracy of your figures.
posted by Jimmy Havok at 11:20 AM on February 28, 2010

Mmm, yeah...can I get in on that 10% a year thing you've got going? In other words,you pulled that unrealistically high number out of...well, it wasn't anyplace reputable.

Come on, don't do this. As I'm sure you well know, that number is just for the purposes of illustration. I typed "MARR" into Wikipedia and it said that most companies use 12%, so I picked 10% to be safe, and threw in the words "roughly" and "or so" for good measure. Bear in mind that it has to include inflation plus growth, and inflation is probably around 5%.

Then you have to look at the cost of alternatives to your course of action. Not building non-petroleum energy sources isn't free, since the capacity has to come from somewhere, so you can deduct the cost of building that conventional plant from the cost of the non-petroleum plant, as well as the environmental cost (which is currently hidden pretty well) of that conventional plant.

I thought we were talking about comparing the costs of different sources of energy. In that case you don't subtract the cost of the other sources, because the whole point is to compare them. That's like subtracting the cost of the salad you normally get from the cost of the sandwich you're thinking about getting when comparing their costs.
posted by Xezlec at 1:52 PM on February 28, 2010

I'd like to comment on the idea above that solar modules stop functioning after 25 years. Most manufacturer warrant their products to produce 80% of their original power output after 25 years. There are Spectrolab solar panels from the 1960s out there in the USA still on rooftops producing close to their original power output after 40 years.
posted by thewalrus at 11:57 PM on February 28, 2010

George Monbiot, not known as an enemy of low-carbon energy, comes out swinging today against the UK's new feed-in-tariff plan for solar.
posted by Dasein at 5:52 PM on March 1, 2010

So we have a 60c a kWh FIT now in NSW, Australia. At my house, over a year, a 1kW panel outputs about 5kWh per day. Four or five year return on investment (less because there are also some federal rebates).
The cheapest energy is still from conservation - CFL lights alone cut 20% of my bill, but I will be installing 3kW of PV to cover about 75% of my usage.
Just be aware of Jevons paradox, and don't install that new bigger refrigerator now you have extra power.
posted by bystander at 4:21 AM on March 2, 2010

Caltech Makes Flexible, 15%-20% Efficient Solar Arrays: "Instead of being flat-panel, they stand thin silicon wires in a plastic substrate that scatters the light onto them. The total composition is 98% plastic, 2% wire — the amount of silicon used is 1/50th that of ordinary panels. So as soon as they can get these to market, solar could be very viable and cheap to produce."
posted by kliuless at 8:16 AM on March 7, 2010

In yesterday's science news was a story about a super-hydrophobic surface. Essentially, the surface is "hairy." Water beads up and rolls off, taking dust with it.

Make the hairs power-generating, and you have a super-efficient, self-cleaning solar power convertor.
posted by five fresh fish at 9:53 AM on March 7, 2010

There are a number of possibilities for efficiently integrating the output of fluctuating generators like wind and solar to the grid. One that just occurred to me is to use the output of a wind farm to pump water into a reservoir that feeds a hydro-electric generator.

Anyway, I believe the grid feed issues are just growing pains, and that there will be a number of ways to better handle this.
posted by Artful Codger at 10:01 AM on March 7, 2010

New "Hairy" Material Is Almost Perfectly Hydrophobic
posted by kliuless at 10:10 AM on March 7, 2010