The age of cheap solar energy – now, to be precise.
by Vinay Gupta • July 26, 2008 • The Global Picture • 8 Comments
Energy is about to become absurdly cheap at least during daylight hours in sunny areas of the planet.
http://nanosolar.com is retailing panels (maybe a gigawatt a year) for $1 per watt. Their production cost is 30 cents per watt of panel capacity. Konarka claims 10 cents per watt production costs (confirmed here) and are expecting to begin mass production in a year or two.
So what does this mean in terms of electricity supply? Simply put, it means that in some applications, solar power’s real cost is about half that of a coal fired power plant today and it’s only going to get cheaper. We’re likely to see solar displace nearly all of the world’s coal plants within 20 years, cutting CO2 emissions by 40%.
Would you like to see the numbers?
Let’s look at the optimal case: direct drive solar panels being sold at the around the cost of manufacture. Direct drive simply means there are no batteries or chargers in the system: the devices being powered are directly connected to the solar panels, so there is very little waste an no additional system costs. This is the sort of setup you might see in a solar powered factory that only runs in daylight hours. The low profit margin is typical of commodity goods like paper or corn. A very large global solar market will tend to push prices down, but perhaps not this close to manufacturing cost. But it does tell us how much it really costs to do, so it’s a good place to start thinking about the numbers.
So how much will power cost in that configuration?
$0.30 per watt of panels means $300 for 1 kilowatt of panels.
20 years of panel life * 365 days per year * 8 hours a day of sun = 58400 hours of electricity generated by the panels over their service lifetime
cost per kilowatt per hour = $300 / 58400 = 0.51 cents per kilowatt hour
Right now, the cheapest coal power in America costs 4.63 cents per kilowatt hour. In this configuration, solar power is 1/8th of the cost of grid power from coal plants.
Now this is not an apples to apples comparison. Our notional power plant is sitting in the middle of a desert area directly driving machines which only need to run in daylight. There are no grid costs, staffing costs, costs-of-borrowing or other factors calculated in. We have not accounted for clouds. But even when you add those factors back in, the conclusion is the same: in real terms, even when you factor in all of those costs, solar power is likely to produce power at about half the cost of the cheapest coal-fired power plants.
The implication is clear: the global power economy is in the early stages of a shift as profound as the shift when gasoline began to be widely used. With new battery technology expected in the near future the issues around powering systems at night and handling spikes in demand my begin to be solved, resulting in a complete green power revolution.
Cost per kilowatt hour for solar panels over 20 year life of panels should be $.0051 not $.51 per kilowatt hour as stated
We need to look at the stuff the panels are made of. No rare metals or toxic ingredients.
That said, it’s good that there’s competition between similar technologies, and also different technologies. The one-size-fits-all, fossil-fuel-or-nothing mentality is so outdated!
And, while we’re at it, does converting at least some of our gadgets to solar kindly force some redesign of the gadgets themselves? Like 12-volt freezers or something like that? That kind of redesign, brought about because no high-voltage energy transport is needed, would in itself change quite a few things, I’d like to guess.
Just some quick thoughts:
1. We need to be very careful with PV industry manufacturing cost estimates. In order to drive up capital many times they are a tad over-optimistic. In Nanosolar’s case for example, they still have considerable work to do on encapsulation. In general it is quite challenging to ramp up lab scale to MW scale (to say nothing of GW) production. That said – many more “conventional” thin film technologies (e.g. amorphous silicon) should be able to easily produce solar PV at costs <$1/W using proven technology once they scale to true mass production .
2. As the demand for solar anywhere near $1/W at the GW level will be ~800X the supply in the US alone — we will not be able to buy the cells at that cost but for whatever the demand bids it up to. You can see this effect now with First Solar’s cells being sold significantly higher than their production costs.
In general I would agree with the conclusions: coal is on the way out – the cost savings are clear with larger plants and better technologies in solar. Direct drive solar is interesting — although I would be tempted to target thermal processing first because of the efficiency advantages.
I would like to see a bit more sophisticated analysis comparing PV direct drive to more complex systems with storage or coupled to cogen….I think the economics may work out even better as you could maintain production around the clock – which should increase your ROI.
Production can be scaled cheaply: Nanosolar quoted $160 million for a 1GW per year panel printer. That means they can do exponential scaling *easily* with a financial turn around of less than a year assuming $1 / watt retail and raw materials of about 30 cents.
You can see how wide those margins are and how fast this might all happen. We’ll see, but I’ve basically stopped worrying not out of optimism, but out of willingness to concede that further analysis is likely to make a more accurate guess about *when* the problem goes away, but not going to change the conclusion that it *is* going to go away.
It might be 5 years, it might be 20, but it’s not going to be never, and that’s the critical bit in terms of understanding future scenarios from my perspective.
For all kinds of reasons, 20 years might be too late.
Vinay,
I would took with a BIG grain of salt everything
Nanosolar or its CEO Martin Roscheisen says about
their panels, production or prices.
I.e. I don’t believe a word before I have myself
purchased one of their panels from an honest retailer
(whenever they will be available, if ever?),
and measured its performance with my own multimeter.
See e.g.
http://greenlight.greentechmedia.com/2009/03/16/will-the-real-cigs-vendors-please-stand-up-1224/
and some of the comments at:
http://www.greentechmedia.com/articles/feeding-in-renewable-energy-breakthroughs-5556.html
Sometimes I think the whole CIGS-hype is yet another ponzi-scheme with venture money.
Notwithstanding, I likewise believe that eventually
Solar Energy will be very cheap, although probably
not with the products of the guys currently making
the loudest claims.
Check this list
http://www.greentechmedia.com/articles/150-solar-startups-part-1–5286.html
for a comprehensive survey of start-ups in this field.
CSIRO, Australia, have recently announced that they are close to developing printable solar panels using polymer technology – using similar technology for printing polymer money.
When up and running they said that they should be able to print enough panels every two months to equal the power output of a coal power station. With power produced at a cost of 50c (AUS) per watt.
This printable solar PV is getting close to being seriously competetive.
How do these new panels hold up under heat? Solar is not maintenance-free, they have to be cleaned, and generally do not produce their best results when they get too hot. Where the best sun is, it tends to get quite hot. Most plastics don’t handle the stress of UV exposure for 20 years.
I’m optimistic these challenges will be addressed, especially for small-scale solar (homestead), but on the large-scale there are already problems, as noted by the NYT: http://www.nytimes.com/2009/09/30/business/energy-environment/30water.html?_r=2&partner=rss&emc=rss