More recent events in solar power

March 3, 2011 1:56 PM

Around one year ago we saw some of the recent events in solar power. At that time solar panels topped out at a peak efficiency of around 290W for a 1.99 x 0.99 meter 72-cell module, with a lone rare and expensive 315W module that was used to build team Germany's solar decathlon winning house. Since then prices have dropped a lot, and China is advancing in commodity tech.

About a year ago the best modules that were available at "normal" prices would output about 135W/m^2 under standard test conditions.

The main event is that commodity cell manufacturers in China are now producing 156mm cells consistently above 18% efficiency. Here's another example of a a generic manufcturer producing 156mm cells with >18% efficiency and >4.35W per cell. Unlike the Sunpower module which is a weird non standard size, all of the high powered modules from China use 72 cells and generally measure 1.99 x 0.99 meters, which makes mounting easy. Generally everything is the same width, 95 to 99mm, with 60-cell modules measuring 1.65 meters long. The peak power output for a really good 60-cell module is now around 265W, equivalent to 162W/m^2. This advance in power density per cell means that what was a 135W/m^2 product one year ago may now be 155W/m^2 in the same size, and consistently sub $1.85/watt.

This means it's now possible to build standard sized panels which use 72 x 156mm cells with 300W to 320W and even 325W ratings. Casual trolling of manufacturers and dealers posting on Alibaba and its competitors shows that prices have stabilized below $1.80/watt for 210W to 300W modules, about a 20% drop from last year. On the US domestic front not all things are going so well, with a major US panel manufacturer moving its operations to China and reneging on factory startup credits provided by the state of MA.

Exotic triple junction gallium arsenide based cells used for space applications exceed 30% efficiency, but they're generally only seen on spacecraft or one-off handbuilt solar race cars in which cost is not a factor.

posted by thewalrus (80 comments total)

Size and efficiency really aren't the main game in most applications, most of us have enough roof space, it's dollars per watt that really matter. Still a loooong way from being competitive however, I wouldn't bet much money on PV as the technology of the medium term future. Far better bets in solar thermal for the next while, and in the long term, geothermal, forth generation nuclear and wave tech.

But anyway, it's still good news.

Evergreen shutting down and moving to China is good news for Americans, even if you don't quite understand it.
posted by wilful at 2:05 PM on March 3, 2011

Evergreen shutting down and moving to China is good news for Americans, even if you don't quite understand it.

can you elaborate on that point?
posted by Dr. Twist at 2:12 PM on March 3, 2011 [1 favorite]

Evergreen shutting down and moving to China is good news for Americans, even if you don't quite understand it.

Can you explain that statement?
posted by dibblda at 2:12 PM on March 3, 2011 [1 favorite]

jinx
posted by dibblda at 2:12 PM on March 3, 2011 [1 favorite]

We're in the middle of PV rush where I live in SW England, as the UK Government initiated a high tariff payment for PV under 5MW (i.e. all of it in the UK). The UK Government has now brought forward the review of the tariff programme that was supposed to happen in 2013 because it looks like they are going to get far more PV than they bargained for and it will add up to a fair amount. Anyway, I mention this as I had a potential developer on the phone the other day and he reckons the are being offered deals on multi-MW installation at ~£1000/kW installed, which would be under US$1.60/W.

This is also pretty interesting to me because it tends to make things look like Germany's big gamble of putting lots of funds into PV expansion by moving early on a high PV tariff payout in order to provide a solid market demand for its PV manufacturers is not going to pay off. China looks like it is going to take over the market before Germany has had a chance to milk the growing international market.

While China may end up with other renewable energy technology markets, I think there is more proof against total capture, since most RE tech is not as modular as PV. This is interesting because a clear case can be made that many of the big national investors in RE tech have done so not just for environmental or even security of supply reasons but also to capture industrial market share, as with Denmark, Germany and Spain for wind turbines. If China is going to keep moving on and growing to dominate emerging renewable energy technology markets if could well dampen some enthusiasm in the EU and US.
posted by biffa at 2:16 PM on March 3, 2011

Evergreen shutting down and moving to China is good news for Americans, even if you don't quite understand it.

Explain it for the Americans who can't under stand.
posted by rough ashlar at 2:17 PM on March 3, 2011 [1 favorite]

...peak efficiency of around 290W for a 1.99 x 0.99 meter 72-cell module...

Neither watts, meters nor cells is a unit of efficiency. A 2 m² area should get around 2kW, so this is in the area of 15% efficiency.
posted by DU at 2:17 PM on March 3, 2011 [1 favorite]

China has a competitive advantage in making solar cells. American resources are being tied up in something they're not good at, they should be doing something they have a relative advantage in. And the subsidies used to prop open the plant are a direct take from other programs that the government could be supporting.

Economics 101.
posted by wilful at 2:20 PM on March 3, 2011 [4 favorites]

I'm personally about ready to try my hand and building my own solar panels with some of the solar cells that show up on eBay, now under $0.50/watt. I'm sure I can solder and not electrocute myself, right?
posted by Mister Fabulous at 2:21 PM on March 3, 2011

Watts are measured at an instant in time under standard test conditions AM1.5, on the DC output of the solar module. Of course your house's roof is unlikely to ever see 1000W/M^2 and the same spectrum that an AM1.5 "flasher" produces to test and rate a module. The output of the test flasher in the factory determines whether the sticker that goes on the back of a module says it's a 280W, 285W, 290W, 300W or whatever, as there's manufacturing variances in any randomly chosen set of 72 cells soldered together.

A measurement of efficiency is cumulative kWh produced per month.
posted by thewalrus at 2:21 PM on March 3, 2011

American resources are being tied up in something they're not good at,

And how, exactly, are you coming up with this analysis of "not good at"?
posted by rough ashlar at 2:23 PM on March 3, 2011

it's dollars per watt that really matter. Still a loooong way from being competitive

Right. If it's $1.80 per watt — $2.50 installed, let's say, and I can get 2000 hours of sunshine on it (in the desert, maybe), that's 2 KW or $.30 worth of electricity, or an 8-year payback. In the desert. Up here in New England where it snows on the panels to boot, it's probably 15 years. Solar thermal comes in at maybe 8-10 years. Only a good investment if it makes you feel good, too.
posted by beagle at 2:24 PM on March 3, 2011 [1 favorite]

I think he means cheap manufacturing, but I could be wrong
posted by Hoopo at 2:24 PM on March 3, 2011

Only a good investment if it makes you feel good, too.

The only way to get people to switch to sustainable energy is to actually charge them what the unsustainable ones cost. Or at least to point out all the taxes and wars they are paying for to get that "cheap" oil.
posted by DU at 2:26 PM on March 3, 2011 [12 favorites]

Only a good investment if it makes you feel good, too.

or you can handle a time horizon longer than 8 years.
posted by Dr. Twist at 2:27 PM on March 3, 2011 [1 favorite]

I forgot: There is no way on earth that you are going to come anywhere near 2kW per two square meters, unless some major advance in silicon cells occurs. If you have a module using 72 x 4.45W cells, and that module measures 1.95 x 0.99m, you have 166 watts per square meter when the module is being perfectly illuminated with conditions identical to a factory solar module testing sun simulator, or "flasher". That would be 332W for two square meters.
posted by thewalrus at 2:27 PM on March 3, 2011

Still a loooong way from being competitive however...
I don't understand what this means. Does it mean that compared to subsidized oil, unsubsidized solar cannot compete? Or compared to subsidized hydroelectric?

I guess Economics 101 is beyond me. But wilful, it would really help this thread if you would say what you mean instead of making us guess.
posted by Killick at 2:28 PM on March 3, 2011

I believe he means a classic economics sense of comparative advantage, but I could be wrong.
posted by norm at 2:28 PM on March 3, 2011

Yes, I kind of agree with wilful, even though I don't want to. The return on investment for solar panels is really low... it takes tens of years to pay off the the initial investment. Cheaper panels == more people more likely to use them. Unfortunately, due to economics 101, cheaper panels also == move manufacturing to China.
posted by special agent conrad uno at 2:29 PM on March 3, 2011

I think he means cheap manufacturing

Being "cheap" does not always mean "good".
posted by rough ashlar at 2:29 PM on March 3, 2011

Does anyone have an analysis of what the cost of a CCGT w/ carbon cost priced at a level that implies an aggregate reduction in existing atmospheric CO^2 levels? Lets say at $5 bucks an MCF for gas?
posted by JPD at 2:31 PM on March 3, 2011

That's some irony. China, belching out coal power to produce our environmentally friendly solar panesl.
posted by mullingitover at 2:32 PM on March 3, 2011

They're also building 28 more nuclear reactors
posted by thewalrus at 2:33 PM on March 3, 2011

Being "cheap" does not always mean "good"

Solar cells are becoming a "commodity technology" much like other tech components: capacitors, general use chips, PCB LEDs, etc. They are built to a standard and due to materials, labor, waste cost, regulation, etc. it's nearly impossible to manufacture in the US at a competitive price on a global market. The very thing that has made the US competitive in any manufacturing has been the "X factor" of US manufacturing where the US makes items that are difficult/impossible to make or move globally.
posted by Mister Fabulous at 2:34 PM on March 3, 2011

Being "cheap" does not always mean "good".

Nor does expensive. Hit a certain amount of KwH per dollar spent or go home is the game here. Location of production isn't a huge issue for quality, but is for the cost.
posted by jaduncan at 2:36 PM on March 3, 2011

The US can't make solar modules competitively for the same reason the US can't make $80 PC motherboards, $90 19" LCD monitors or $25 802.11n routers competitively. It's all going to come from Asia unless something dramatic changes with US labor rates and factory startup/operation costs.
posted by thewalrus at 2:36 PM on March 3, 2011

They're also building 28 more nuclear reactors

Good thing they think ahead and add lead to the paint on childern's toys. Those kids will have Brain shrinkage protection!
posted by rough ashlar at 2:36 PM on March 3, 2011

The US can't make solar modules competitively

But that has little to do with a statement of "not good at".
posted by rough ashlar at 2:38 PM on March 3, 2011

But that has little to do with a statement of "not good at".

To some extent it does. Can't make them efficiently, if you like. It doesn't really matter if the inefficiency comes in mechanical process or labour costs. If you want a certain result, China can do it using less resources.
posted by jaduncan at 2:40 PM on March 3, 2011

Size and efficiency really aren't the main game in most applications, most of us have enough roof space, it's dollars per watt that really matter.

Although not nearly as big a market, solar panels on camping trailers or RVs represent an extremely useful application and would benefit most in this case from the highest possible efficiency for their weight/size. Provided said camper is equipped with LED lighting and a pair of robust 6 volt golf cart batteries, one could be self sustaining indefinitely with a 300 watt panel, and that's even with regular stereo or television use.

To use a furnace fan, air conditioner or microwave, however, the solar panel investment would have to be quadrupled, or more, depending on your location and the size of the camper.
posted by CynicalKnight at 2:41 PM on March 3, 2011

The US is still "good at" making incredibly complex high-tech stuff, there's an IBM chip fab somewhere in upstate New York that makes microprocessors for defense contractors... The sort of application where the pluggable module holding the keys on your the Type 1 crypto thing plugged into your Harris manpack radio absolutely, positively cannot ever be touched by a non-US citizen. It's a boutique market.
posted by thewalrus at 2:41 PM on March 3, 2011 [1 favorite]

CynicalKnight, another market that cares a lot about efficiency in a relatively limited space is sailboats. The US has a lot of privately owned boats and people love to put solar panels on their sailboats. What good is a 45' long living space equivalent in area to a small one bedroom apartment if you can't use all your electronic gadgets?
posted by thewalrus at 2:42 PM on March 3, 2011

I've always thought it'd be neat to have a few solar cells on cars to recharge batteries and electronics, in case I ever turned off interstate at the wrong place and stranded myself on a logging road.

And on preview I guess you might want to get far away from civilization intentionally.
posted by pwnguin at 2:49 PM on March 3, 2011

Out of curiosity, what's the number you get when you divide the cost per watt of a solar pannel (let's say over its lifetime) by the cost per watt from coal at a modern power plant? Or it divided by the cost per watt of electricity from a nuclear plant over its lifetime, including the cost of construction and engineers to run the thing?

I know they say solar won't catch on outside of environmentalists and people off the grid until it's cheaper than fossil fuels, but I'm curious how far we are from that point. I've heard some people (I think Ray Kurzweil) say that Moore's law applies to solar panels, but that seems kind of fishy to me that it would apply to two disparate fields.
posted by mccarty.tim at 2:51 PM on March 3, 2011

I really doubt anything like moore's law, the progession of efficiency from 15% on monocrystalline cells to 18% on the same medium took many years. the manufacture of a slab of monocrystalline silicon to produce some electricity when sun hits it is nothing like the intricate design work of 32nm multi-core CPUs on a single die.
posted by thewalrus at 2:53 PM on March 3, 2011

So... power storage has caught up, right? ... right?

Suppressed nanotech, what?
posted by vrogy at 2:56 PM on March 3, 2011

But that has little to do with a statement of "not good at".

I took it to mean "not good at producing these things cheaply, so that people will be able to/be tempted to buy them." Which is important. I highly doubt it was a swipe at the competence of the American manufacturing sector. Jeez, so sensitive!
posted by Hoopo at 2:56 PM on March 3, 2011 [1 favorite]

Responding primarily to wilful: it is problematic to apply classic comparative advantage analysis to fields, such as high technology (advanced manufacturing, etc) that do not relate to country's inherent capital stock or natural resources. Long story short, one country or another can learn to become better at making solar cells quickly, cheaply, and efficiently.

The classic illustration of comparative advantage is as follows. Assume 1 unit of wool is worth as much as 1 unit of wine. The argument is that England can make 1 unit of wool in 10 person-hours, and 1 unit of wine in 15 person-hours. Portugal can make 1 unit of wool in 8 person hours, and 1 unit of wine in 6 person hours. Even though it's ostensibly "cheaper" for Portugal to make both domestically, in fact Portugal should just focus on making wine and trade it to England to get all the wool they need, even though it's marginally more expensive to make wool in England.

Underlying this, however, are the implicit facts on the ground that England has lots of nice rolling pastures to raise sheep, and Portugal has plenty of sunny weather to raise their vineyards - part of what's striking about this example is it actually tracks history fairly well, though of course it is oversimple.

The problem in applying this to the high-tech arena is that no one nation is inherently better at making or using the machines used to manufacture the goods. Sure Chinese labor is cheaper per raw person-hour, but it is non-intuitive that this is really the limiting factor in the cost picture, where getting good quality relies at least in part on, e.g., the application of sophisticated science education. Or at least it does if you're interest in re-investing gains of innovation into improvements of the process.
posted by rkent at 2:57 PM on March 3, 2011 [1 favorite]

The US can't make solar modules competitively for the same reason the US can't make $80 PC motherboards, $90 19" LCD monitors or $25 802.11n routers competitively. It's all going to come from Asia unless something dramatic changes with US labor rates and factory startup/operation costs.

Of coruse. Which is why the world's No. 1 solar cell and panel manufacturer as of 2009, First Solar, did its early R&D in Toledo, Ohio, headquartered itself in Phoenix, Arizona, and built its first large-scale manufacturing plant overseas - in eastern Germany.

Labour's not the biggest variable in a solar panel's cost. The only reason the Chinese panels are cheap is because the Chinese government has so thoroughly subsidized their production that they're coming to market at below production cost. Key line from that last link:

Shi Zhengrong, the chief executive and founder of China’s biggest solar panel manufacturer, Suntech Power Holdings, said in an interview here that Suntech, to build market share, is selling solar panels on the American market for less than the cost of the materials, assembly and shipping.

When I was reporting on this stuff in Germany in 2009, this admission, apparently a slip-up, in the Times, was the talk of the industry. This is, in part, about who wants pole position in the cleantech revolution. The US has among the worst regulatory environments for renewables in the industrial West, China and Germany among the best.
posted by gompa at 2:59 PM on March 3, 2011 [12 favorites]

mccarty.tim :

Have a look at the entries on Barry Brook's website about his peer reviewed paper in Energy on comparing nuclear energy to solar power cost.

He also has an interesting comparison between Denmark and France's change to their energy system.
posted by sien at 3:00 PM on March 3, 2011

I'm pretty sure the First Solar factory was in Germany to take advantage of the German government's domestic content requirements for their feed-in tariff. Since it was introduced something like a dozen amorphous silicon thin film manufacturers set up shop in germany printing Si onto big sheets of glass. Many of them use the same factory equipment. First Solar is unique, but the rest of the german factories are all very similar.
posted by thewalrus at 3:03 PM on March 3, 2011

Shi Zhengrong, the chief executive and founder of China’s biggest solar panel manufacturer, Suntech Power Holdings, said in an interview here that Suntech, to build market share, is selling solar panels on the American market for less than the cost of the materials, assembly and shipping.

Sounds a lot like how China cornered the rare earth market and then started to throw it's weight around when it succeeded.

So are we going to have a replay of the same scheme with the same results? Probably.

What was all that talk about energy independence again?
posted by dibblda at 3:05 PM on March 3, 2011

I'm pretty sure the First Solar factory was in Germany to take advantage of the German government's domestic content requirements for their feed-in tariff.

No, I'm absolutely certain there are no domestic content requirements in the German FIT, and that the factory is there because there were really good tax incentives in place to employ skilled workers in eastern Germany and because it was close to the biggest PV market in the world at the time. Germany itself, that is.

I could go linkhunting, but I've interviewed First Solar's current head of German operations and both the lead authors of Germany's FIT, so trust me on this one.
posted by gompa at 3:08 PM on March 3, 2011 [2 favorites]

"I've heard some people (I think Ray Kurzweil) say that Moore's law applies to solar panels, but that seems kind of fishy to me that it would apply to two disparate fields."

Moore's law says we can find ways to make smaller transistors, in essence. Which they've done for quite some time. Solar cells, while semiconductors, are more concerned with finding new chemistries that capture a broader set of wavelengths. Different challenges, and the history shows a different growth pattern. For the most part, we see jumps and slumps in growth.
posted by pwnguin at 3:14 PM on March 3, 2011

Does any of this mean I can charge my laptop with solar panels on a road trip...for cheap?
posted by jnnla at 3:14 PM on March 3, 2011

Does any of this mean I can charge my laptop with solar panels on a road trip...for cheap?

Depending on your definition of "cheap," probably you can.
posted by Kirth Gerson at 3:30 PM on March 3, 2011

Or it divided by the cost per watt of electricity from a nuclear plant over its lifetime, including the cost of construction and engineers to run the thing?

And the cost of getting rid of the waste? Rookie mistake.
posted by Mental Wimp at 3:36 PM on March 3, 2011 [2 favorites]

The only way to get people to switch to sustainable energy is to actually charge them what the unsustainable ones cost. Or at least to point out all the taxes and wars they are paying for to get that "cheap" oil.

I agree with you in principle that negative externalities should be passed along to consumers in the form of a tax. However, Solar power on its own will never have a measurable impact on oil consumption. It will mainly make a small nick in our reliance on coal and (to a lesser extent) natural gas.
posted by schmod at 3:46 PM on March 3, 2011 [1 favorite]

It's all going to come from Asia unless something dramatic changes with US labor rates and factory startup/operation costs.

Out of curiosity, how much of a semiconductor fab's startup and operating expenses come from labor costs? I always assumed that those places would be mostly automated, and therefore not hugely influenced by labor costs.
posted by schmod at 3:47 PM on March 3, 2011

I'm personally about ready to try my hand and building my own solar panels with some of the solar cells that show up on eBay, now under $0.50/watt. I'm sure I can solder and not electrocute myself, right?

Not as easy to solder as you might think.. Also, the choice of leading wire is a lot more sensitive than you think.

Consider, if you use lead bearing solder you can get away with heating the wire to a little less than 200F, but that is still enough heat to make the copper expand by ~20%. The cells don't expand anything like that much, so when you are done the cells are under quite a bit of tension. Use non-lead solder and the heat goes up, making even more stress.

Also, it isn't that easy to solder to the contacts on the cells. I found it was easier to apply solder paste with a makeshift silkscreen, then apply wire, and reflow with a hot air station. Using lead bearing solder paste is probably extremely bad for your health, because solder paste gets everywhere.

Still totally doable, don't get me wrong.
posted by Chuckles at 3:55 PM on March 3, 2011 [1 favorite]

Out of curiosity, how much of a semiconductor fab's startup and operating expenses come from labor costs? I always assumed that those places would be mostly automated, and therefore not hugely influenced by labor costs.

my impression is that it isn't the fab cost that differentiate costs - as you say its really just a lot of capital + electricity. Its all the other steps in the process where labor matters. Also as other have alluded to - subsidies both direct and indirect.

Anyone have an answer to what carbon costs if we force the world to roll back CO2 to a sustainable level? And then what that implies for the economics of gas fired generation vs solar?
posted by JPD at 4:02 PM on March 3, 2011

Still totally doable, don't get me wrong.

I think it's about time to burn my AskMe question for the week.
posted by Mister Fabulous at 4:13 PM on March 3, 2011

Given the US federal tax subsidies for solar power, it just barely makes sense if your are paying the national average for electricity and assuming $4/Watt installed. However, I'm pretty sure that calculation fails to take into account the steady decline in efficiency over the life of the panel (add 1/27 total system costs every year?). Given no subsidies it's a system that never pays off. For that reason, if I owned my own home I'd install solar, but I think it's bad policy for the government to blindly subsidize it (high electricity cost states like CA may be an exception).

I think we'd be better off focusing on municipal installations, and new housing installations. Ideally with the solar panel being integrated into some kind of roofing system so that it's directly replacing part or all of the expenses of roofing a new home. This patchwork retrofit install on older homes just doesn't seem like a very efficient way to drive solar adoption, even if it does help with localized peak grid demand in areas where people use their AC in the summer.

The solar calculator I used has some flaws, but I'm too busy to make a better one.
posted by BrotherCaine at 4:14 PM on March 3, 2011

people suggesting I was taking a potshot at good ole US quality have got the wrong end of the stick. Mind you, without checking I guess some of those people were taking a potshot at chinese quality, which is equally false. Americans are really insecure about these issues generally I have observed. There's a lot of latent sinophobia on metafilter.

(As an Australian, I've accepted that we're bloody good miners, farmers and tourism service providers, and not much else.)

The money that Boston invested in this company is money that ought to have gone to programs that the government has a role in, that have a market failure. You're picking winner and subsidising a company that can't turn a buck. The money was wasted and that was inevitable. It's good that the company moved to China before taking away more taxpayers money. The argument that China maybe in fact doesn't have an advantage runs into the empirical fact that this company is moving there.

Whether or not China is subsidising solar manufacturers isn't proven, but if they are, the answer isn't to subsidise it yourself!

Oh, by the way, I'm getting more solar panels on my house in a few weeks (having installed the first lot in 2006). But that's only due to government subsidies that are morally wrong. The dollar per tonne of carbon saved is something in the order of $200. These panels are cheaper than the ones I had five years ago, and they're made in China. Yay for me! Yay for the Chinese! Boo for the Australian government!
posted by wilful at 4:21 PM on March 3, 2011 [2 favorites]

Oh, by the way, I'm getting more solar panels on my house in a few weeks (having installed the first lot in 2006). But that's only due to government subsidies that are morally wrong. The dollar per tonne of carbon saved is something in the order of $200. These panels are cheaper than the ones I had five years ago, and they're made in China. Yay for me! Yay for the Chinese! Boo for the Australian government!

But its overly simplistic to consider subsidies for non-mature technologies only in terms of their current costs. It is quite likely that those panels wouldn’t be so much cheaper if there hadn't been subsidies elsewhere, most notably Germany. Investing in less mature technologies pushes them along their cost/experience curves such that they become cheaper, and offers clear public good benefits, including enhanced security of supply, reduced reliance on imports and reduced environmental impact. Obviously Australia is not likely to capture significant elements of the industry, and some countries (Germany, Japan and the US have been the three nations with the longest running focus on PV - not unconnected to their existing semiconductor sectors) have taken a gamble on capturing market segment and it’s not looking good that they will now be able to stop China from coming in and seizing market share. This has come as a pretty big surprise in the long term (and national industrial policy runs at least a decade ahead), it’s not like there was any reason to think China would jump ahead in a sector like this. If you compare this with the medium tech wind sector, Denmark has more than seen a return on the money it spent to keep its turbine manufacturers going through tough times in the 1980s, and keeping domestic demand going through the 90s. China has made in-roads there too but is far from dominant.
posted by biffa at 4:37 PM on March 3, 2011

BrotherCaine writes "I think we'd be better off focusing on municipal installations, and new housing installations. Ideally with the solar panel being integrated into some kind of roofing system so that it's directly replacing part or all of the expenses of roofing a new home."

Re shingling as well. Even if you had to lay down a layer of very cheap rolled roofing for your waterproofing and just used the cells as a UV barrier the cost savings over expensive 25 year shingles would be significant, attic temperatures would be lower reducing A/C loads and your roof would last longer. Especially in places like Arizona, New Mexico, or Nevada where roof surface temperatures are high for long periods of time.

I'd also like to see someone develop a system for flat roof buildings that would allow you to walk under the cells there by shading the roof while still allowing for maintenance. We'd see some energy savings from higher A/C efficiency if over all roof temperatures were reduced. Not a lot mind you but 1% of a 100 tons of air conditioning would be noticeable.
posted by Mitheral at 4:39 PM on March 3, 2011 [2 favorites]

Thankfully; China signed the Kyoto Protocols.
posted by buzzman at 4:49 PM on March 3, 2011

Investing in less mature technologies pushes them along their cost/experience curves such that they become cheaper, and offers clear public good benefits, including enhanced security of supply, reduced reliance on imports and reduced environmental impact.

Investing in research, definitely. I don't know enough about how well production subsidies in solar pay off to speculate, but I'd be curious to hear from someone who does have expertise.
posted by BrotherCaine at 4:52 PM on March 3, 2011

I doubt BrotherCaine's estimate of PV panel degradation is realistic, modern panels do notdegrade at greater than 1% per year. The US DoE NREL in Colorado has a number of polycrystalline panels installed on a long term outdoor testbed that were put in place in the mid 1990s, over 15 years they've degraded less than 5%. Here's one example showing degradation of amorphous Si thin film modules after their initial "burn in " power drop. Most major manufacturers warrant their modules to produce 80% of their original power output after 25 years and that is an extremely pessimistic figure compared to what will actually happen.
posted by thewalrus at 5:36 PM on March 3, 2011 [1 favorite]

China has a competitive advantage in making ________. American resources are being tied up in something they're not good at, they should be doing something they have a relative advantage in. (i.e. Eating snack foods in front of the television.)

The subsidies used to prop open the plant are a direct take from other programs that the government could be supporting. (Besides, they're no where near the level of China's subsidies for making _________.)

China invests in its future. The US steals from its people's future in exchange for tax breaks focused on the rich and a military that spends more than the whole rest of the world combined. So Americans might as well just give up, their asses down on the couch, order a pizza, and watch some Glee.

Economics 101.
posted by markkraft at 5:55 PM on March 3, 2011 [2 favorites]

Thanks for the link thewalrus, I had seen out of date information with a much higher drop in efficiency over time (perhaps the Tunisian study referenced in your link 5% drop/yr due to "encapsulation browning"). That combined with what I was told by solar system vendors in the past made me assume there were some non-negligible maintenance/replacement costs. I'm glad they've improved things.
posted by BrotherCaine at 6:14 PM on March 3, 2011

They're also building 28 more nuclear reactors

Hmmm. I wonder which way the wind will be blowing when they have their Chernobyl.
posted by Twang at 7:10 PM on March 3, 2011

There's a lot of latent sinophobia on metafilter.

Latent?
posted by pompomtom at 7:59 PM on March 3, 2011 [2 favorites]

non-negligible maintenance/replacement costs

I haven't read thewalrus' links, but I assume those installations do have to clean surfaces once in a while to keep performance up. So ya, non-negligible maintenance for sure, but not frequent replacement.
posted by Chuckles at 8:25 PM on March 3, 2011

Today's standard solar panels are not high tech, they are not significantly different from panels manufactured 20 years ago though the materials, efficiency and reliability are all much improved. The manufacturing process is basically the same.

The whole point of trying to get utility scale solar PV, means driving every bit of the supply chain to rock bottom commodity prices. The USA was never the right place for the manufacture of standard solar panels in the long term. State and local governments should probably pay attention to that when they're playing VC with taxpayer money.

Make no mistake, China is subsidizing its solar business along with many others. However I expect just like Japan and Korea subsidized their semiconductor business back in the 80's it probably won't pay off for them as much as they hope. Those manufacturing jobs moving to China this decade will be off to Vietnam and the next tier of low cost labor developing nations before long and that's a good thing. When your labor costs have risen to the point where the shoe(and now solar) manufacturers have left, then a great many people have transitioned out of real poverty to at least a working class existence.

I for one think that if all those Walmart $$ going to China end up subsidizing solar panel manufacturing to the point that its competitive with fossil fuels, then it won't have been so bad. See communism works!

The USA's strength lies in inventing and manufacturing the real high tech stuff and I believe we do it better than just about anyone else. There are new solar technologies being worked on in Silicon Valley that will make those cheap Chinese panels look like expensive low powered toys. There's a lot of reasons to be optimistic.
posted by Long Way To Go at 8:38 PM on March 3, 2011

Ah, yes. Whatever happened to the idealab solar thingy?
posted by maxwelton at 9:27 PM on March 3, 2011

"Hmmm. I wonder which way the wind will be blowing when they have their Chernobyl."

I suspect they never will.... especially given that a great deal has happened to make new nuclear plants considerably safer, and that there's a finite reserve of uranium in suitable concentrations as to make cost-effective energy.

That, and the Chinese tend to be appropriately stern for those who, through corruption or negligence, bring dischord and death. Fatally stern, even.

It must be somewhat daunting to be someone overseeing a big project like a nuclear power plant in China, frankly. On one hand, it seems almost de riguer to embezzle some of the money being spent on the project... but you'd want to make absolutely sure you don't leave your widow paying for the bullet you're executed with later.
posted by markkraft at 9:30 PM on March 3, 2011

Twang writes "Hmmm. I wonder which way the wind will be blowing when they have their Chernobyl."

Do you wonder about the wind direction over France because though China has developed a new reactor in co-operation with Westinghouse the vast majority of their installed base and reactors under construction are of French design.
posted by Mitheral at 9:38 PM on March 3, 2011

Ah, yes. Whatever happened to the idealab solar thingy?

If you mean Idealab eSolar then it looks like they have a 5MW demo plant up and running in Southern California. Not too bad considering they were founded in 2007.
posted by Long Way To Go at 9:43 PM on March 3, 2011

First Solar 2010 Results and 2011 Guidance

2010 sales of $2.564 billion; 11.6% efficiency at $0.75 per-watt cost; 2 gigawatts of production guidance for 2011.

Solar cells are becoming a "commodity technology" much like other tech components: capacitors, general use chips, PCB LEDs, etc. They are built to a standard and due to materials, labor, waste cost, regulation, etc. it's nearly impossible to manufacture in the US at a competitive price on a global market. .

Details are always important when making these types of analyses. Cost of shipping from China (fuel price dependent, solar panels are not tiny chips, PCB or little ipods), factory automation/location, solar cell technology, cost per Watt of electricity from centralized sources in the near future all make this statement shortsighted at best, completely wrong at worst.

This is, in part, about who wants pole position in the cleantech revolution. The US has among the worst regulatory environments for renewables in the industrial West, China and Germany among the best.

Damn right.
I blame US utilities. They fear decentralization (lack of power, heh!) of solar and wind and energy storage, so we get nuclear power loan guanantees, patchwork solar inititives and no federal feed-in tariffs.
posted by peppito at 11:11 PM on March 3, 2011

I'm an electrical engineer working at a solar manufacturing start-up in San Jose, CA. Many of us are skeptical about Chinese solar manufacturers and here's why:

From what I've read and seen about Chinese companies, they mostly make some kind of silicon modules. Crystalline silicon solar cell production requires a lot of money because you must grow silicon ingots, cut them, and then handle them individually throughout their processing. Yingli solar, one of the more promising companies in China, produces polycrystalline silicon cells, a thin film variation sacrificing a little efficiency for a sizable cost savings.

So most Chinese companies apply the standard Chinese manufacturing model of cheap, plentiful labor to the manufacturing process. The solar industry there also benefits from government subsidies in the form of cheap land and loans. But in the long run, as China develops, their labor will cease to be so affordable and their subsidies will dry.

In the US we have a strong hand of companies promising sub dollar-per-watt solar modules. The list starts with FirstSolar (FSLR) which was the first to cross the $1/W threshold. It now employs 4700 people and is slated to produce 2.2GW worth of panels this year. They have a successful process already and simply need to ramp up as fast as possible and rake in the cash.

Many other companies promise the same, and some might follow through: Solyndra, which produces CIGS cells on cylindrical tubes, made a lot of promises and is quickly going defunct. Miasole makes CIGS cells on a thin stainless steel substrate and is currently ramping up production. Stion is still in the start-up phase, working on its manufacturing process. Nanosolar prints CIGS on a thin metal substrate and is ramping production.

There are other companies working on things like solar concentrators or trackers. But I personally don't believe either of these technologies will play a major part in the market. PV concentrators haven't shown much promise in terms of cost reduction (please someone correct me on this if I am mistaken), and the biggest customers aren't going to be homeowners or anyone who is strapped for space, where it might make sense to squeeze out every last watt at the cost of expensive tracking hardware. They will be power companies and utilities who have a lot of land, a huge thirst for power, and no stomach to deal with complicated maintenance on gigawatts worth of panels.

Most of our power currently comes from coal or other plentiful fuel sources. This baseline power is incredibly cheap (the equivalent of pennies/W in solar terms) but is not on-demand. Baseline power plants usually take a long time to start up or stop (with the notable exception of hydro). That means if there is a spike in demand, you cannot react unless you have another power source. This source is called peak power, and is often in the form of a natural gas generator. Natural gas, not other solar companies, is the target competition at the moment. Demand tends to spike on hot, bright days, when photovoltaics are conveniently the most effective.

But back to the Chinese companies: Essentially, they are building a bubble, much as we are in the States. But it is my opinion that when ours pops, due to our greater level of innovation and fewer artificial, temporary crutches, American companies be much more successful than our Chinese counterparts.
posted by cman at 11:44 PM on March 3, 2011 [6 favorites]

I installed 3kW of panels on my Australian roof last year, thanks to combined federal and state subsidies. The state subsidy (NSW gross FIT @ 60c kwh) has since ended for newcomers, as take up was double expectation. The feds are cutting back their subsidies too.
Power here is 20c per kwh, which is something like 4 times what it is in Texas, yet an unsubsidised solar system only breaks even after about 10 years, and if you include time value of money it is touch and go whether it ever does.
Except, it produces no CO2 during operation, and will likely last >25 years, and the time value of money is so low currently it is almost negative. So you get green power for a similar cost to coal power, and any power generated in the far future comes for free.
Thanks to subsidies, I'll likely make a profit on my solar panels within about 5 years, I already get all my power free. My out of pocket was $9000.
Solar is good, in my opinion, in that it is a genuine, concrete step you can make to improve the environment and mitigate peak oil, even if it may not pay you back in dollar terms for some time. Bear in mind power costs are likely to rise faster than inflation, too.
posted by bystander at 4:51 AM on March 4, 2011

Also, RE Moore's law for solar - in a household system a large component of the price is the electrical inverter. It is quite believable that inverter price/performance will increase significantly (although admittedly logarithmic growth would be hard).
posted by bystander at 4:53 AM on March 4, 2011

Point of order: Moore's Law is not, in fact, a law of any kind.

It has about the same scientific force as the Great Eskimo Vocabulary Hoax.

Once the public has decided to accept something as an interesting
fact, it becomes almost impossible to get the acceptance rescinded.
The persistent interestingness and symbolic usefulness overrides any
lack of factuality. -- Geoffrey Pullum

posted by Herodios at 7:59 AM on March 4, 2011

There is indeed a sort of a Moore's law for inverters. Or really, switching power supplies in general. Power densities have gone way up and costs way way down in the last 15 years. I don't really know what projections are on that front, a lot of the things that can be done have been done in the switching power supply business. Those advances haven't hit the inverter business as much as the computer power supply business, but it is coming. Look at how cheap those car plugin 12V inverters are. Whole house inverters are still a bit of a boutique thing, but if/when mass manufacturing gets to them, you can expect around $0.10 per watt.
posted by Chuckles at 8:13 AM on March 4, 2011

Whatever happened to the idealab solar thingy?

The same thing that happened to Dean Kamen's Stirling cycle engine, the Ochicron sub $100 1 HP Nitrogen Stirling, or Catapiller's stirling - it never shipped
posted by rough ashlar at 9:06 AM on March 4, 2011

the wind direction over France ... reactors

Aren't a disproportional number on the French-German border?

Ya don't need a weatherman to know which way the wind blows.
posted by rough ashlar at 9:09 AM on March 4, 2011

cman: China is also ramping up on the amorphous silicon thin film market, with panels that have lower efficiencies but also much lower production costs. There isn't a significant difference in efficiency between monocrystalline silicon and polycrystalline silicon cells from China (156mm). The same 72-cell module that might be 290W using monocrystalline would be 270W using polycrystalline, but the poly module means they can get away with using broken pieces of silicon ingots instead of the whole unblemished thing. I've seen a lot of polycrystalline modules at sub $1.60/watt that had cosmetic issues only, as in the main thing that made the module "ugly" was that some of the cells were significantly lighter or darker coloured than the other, so it had a sort of unevenly distributed checkerboard pattern.

The initial degradation of amorphous si modules is taken into account when rating them. There's a dropoff curve as it "burns in", so if it produces 125W under the tester, it'll get rated as a 108W module. One of the issues that I think the US is going to face is that the CIGS process is much more complex than just printing silicon onto a sheet of glass, which is pretty much what amorphous Si is. US distributors are already buying Amorphous Si modules by the containerload from China for sub $0.90/watt and reselling them from $0.98/watt to $1.19/watt. When the efficiency is sub 100W/m^2 you need a lot of space to mount these, but this works great if you have a huge area to hold panels such as the roof of a warehouse. Then you have to balance out the mounting costs which are not insignificant for hundreds of 1.10 x 1.00m modules...
posted by thewalrus at 9:30 AM on March 4, 2011

I can only assume that some of the installed cost per watt numbers that are being thrown around here are post-government-incentives?

I have a commercial/industrial solar installation business in Chicago, and we just spec'd out a couple medium-size installations (around 500kW each). The solar panels themselves came out to around $2.40/W, and with other parts and labor (union, in this case) the total installed cost was just under $5/W.

That is pretty typical from what I've heard from other contractors; the lowest I've seen on a large commercial job so far is around $3.50/W (that was an easy installation and using cheap-o parts). Parts and labor aren't going to come down too much in the near future, so you're looking at around $1.50-$2.50/W of non-solar-panel expenses. $2.50/W installed will only be feasible in around 2 years, when panels drop below $1/W. And $1.60/W is crazy in the current market - that must be HEAVILY subsidized.

In the US right now, you can reasonably expect to pay under $4/W post-tax-incentive. That's still a hard sell in the midwest, with commercial power costing around $0.08/kWh on average. But on the coasts it's a total no-brainer - at $0.14/kWh or more, with additional asset depreciation incentives, those installations are cashflow positive from day one.

If I plug $0.14/kWh instead of $0.08 into the most recent financial model I did for a customer, it would be a whopping 53.22% IRR over 30 years instead of 8.56% (and that's in Chicago climate - a measly 4.4 hours of sun per day on average.) Compare that to just about any other investment a business could make. It boggles my mind that people in those markets aren't burning up their phone lines trying to get someone out to install solar.
posted by chundo at 4:22 PM on March 4, 2011

chundo: I just checked the final quote I got for my system. Without any government subsidies/incentives it was $17150 for a 2.88kW system installed.
Australian federal gov subsidies (based on paying over the odds for renewable energy certificates) gave a discount of a little over $7700. So about $5.90 per watt without incentives.
posted by bystander at 7:26 PM on March 4, 2011