I guess this post is an excercise in peevishness more than anything else. It bugs me when people who don’t know what the heck they’re talking about are picked up by other people who don’t know what the heck they’re talking about and suddenly that becomes the prevailing wisdom.
What’s bugging me today goes back to an article at Scientific American in which the author asserts that Moore’s Law applies to solar power:
Over the last 30 years, researchers have watched as the price of capturing solar energy has dropped exponentially. There’s now frequent talk of a “Moore’s law” in solar energy. In computing, Moore’s law dictates that the number of components that can be placed on a chip doubles every 18 months. More practically speaking, the amount of computing power you can buy for a dollar has roughly doubled every 18 months, for decades. That’s the reason that the phone in your pocket has thousands of times as much memory and ten times as much processing power as a famed Cray 1 supercomputer, while weighing ounces compared to the Cray’s 10,000 lb bulk, fitting in your pocket rather than a large room, and costing tens or hundreds of dollars rather than tens of millions.
He then goes on to demonstrate with facts, figures, and charts that Moore’s Law does not apply to solar power.
Let’s start at the beginning “Moore’s Law” is a rule of thumb that the number of transistors that can be placed on an integrated circuit doubles roughly every two years. It’s derived from a paper published by Gordon Moore, one of the founders of Intel, back in the 1960s (1965). Moore’s Law predicts quadratic growth in the number of transistors that can be place on an IC. It says nothing whatever about cost and no direct straight-line relationship exists between the number of transistors on an IC and its cost.
The SA article, after asserting a relationship to Moore’s Law, illustrates what may be either an exponential or hyperbolic decline in the price per watt of solar photo-voltaic cells. The author then produces a log scale chart in an attempt at distinguishing between exponential growth and hyperbolic growth and claims, IMO rather unconvincingly, that the growth is exponential. The reason I’m not convinced is that I think he needs to control for the research dollars spent and for the price of oil. What his graphs of price declines in solar photo-voltaic cells look like to my eye are charts of the reciprocal of oil prices.
However, and this is my essential point: quadratic growth (doubling periodically) and exponential growth (increasing at a fixed rate periodically) are not the same thing.
I think there are good reasons to suspect that the returns to investment in solar photo-voltaic cells will diminish, among them limits in how much sunlight falls on a given area. Further, it’s misdirection. Oil still produces significantly more watts per kilogram than SPV and that will continue to be the case for some time. And there are other hurdles. There is no Moore’s Law for batteries, none for energy storage generally.
The SA article was picked up by Paul Krugman who alluded to it in a NYT column and now the idea that there’s a Moore’s Law for SPV is spreading like wildfire.
IMHO you hit all the important points. I’m not sure what the noodleing about quadratic vs exponential is about but you whacked the nail square on the head here: “There is no Moore’s Law for batteries, none for energy storage generally.”
The point being, transistor technology and energy conversion technology are two different things.
The externalities of oil prices, resource allocation and – heh – its cloudy and rainy today in Chicago also apply.
Lastly, Krugman may be a brilliant international economist, but his dabbling in Keynsian economics is suspect…….and now in physics is laughable. That won’t stop him though, nor his sycophants.
Quadratic growth is enormously greater than exponential growth. It’s a difference in kind.
I thought Moore’s Law was actually dead, and has been for some time. The only reason why it still appears to apply is because we keep adding more cores to computers.
In any case, I can’t imagine Solar Power ever becoming the primary source of electrical power in this country. At best, you’re going to end up with a combination of Solar and a secondary power source, probably something that can be ramped up very quickly when sunlight is scarce (such as Natural Gas).
Dave –
Please. I knew that. I’m originally trained as an engineer. Its just that once you decide you are dealing with two different technologies, discussions of growth models are obviated.
Here’s something interesting on the alternative energy front: Taming Unruly Wind Power:
That’s something I’ve been talking about for years, sam. Combine the lack of attention to storage technology (we’ve not particularly good at storing electricity but we’re pretty good at storing heat) with the inadequate attention to the power grid and alternative, distributed energy production isn’t nearly as effective as it might be. The article highlights a novel method of capitalizing on heat storage technology.
Does anyone know why we can’t use large capacitors to store electrical energy? All small electronics do.
Sure. Capacitors won’t hold the charge long enough. It’s something that’s been suggested and they’re working on it.
However, and this is my essential point: quadratic growth (doubling periodically) and exponential growth (increasing at a fixed rate periodically) are not the same thing.
Doubling periodically and increasing at a fixed rate periodically are the same shape of curve, just using different units to describe its curviness. Increasing at a fixed rate of 7% per year (the current trend for solar cells) is the same as doubling every period of just over 10 years, and Moore’s Law (doubling every 18 months) is about a 59% annual growth rate.
On a somewhat different note, actual exponential growth is vanishingly rare in the wild. Usually, what appears to be exponential growth is actually a slice of a logistics curve (or S-Curve), which strongly resembles exponential growth near the beginning (when the main constraint on growth is bootstrapping), then linear growth near the middle (when constraints start balancing out the exponential factors), then asymptotic decay towards the physical limit near the far end (when the constraints come to dominate the growth rate). For example, a cell culture will grow exponentially when the main constraint is the number of cells to make new cells, but will level off as your petri dish fills up, looking linear for a while in the middle.
I’ve made that very point here myself from time to time. I think that exponential growth continuing into the indefinite future is highly unlikely whether you’re talking about economies or cultures in a petri dish (more accurate would be: physically impossible).
The more pertinent question is will growth level off next year? My hypothesis is that when you discount the bubbles growth leveled off a long time ago.
“Sure. Capacitors won’t hold the charge long enough. It’s something that’s been suggested and they’re working on it.”
But Dave…..anyone who had basic high school or lower level college electrical science training knows the issues about capacitors.
If you want to store large amounts of electrical energy you need:
1. big honkin’ plates to soak it up
2. high dielectrics to avoid bleed off.
What I was trying to initiate was a discussion on why these issues haven’t been solved, in the context of your essay.
The engineer in me tells me that the answer is that metals are metals – they have only so much “capacitance.” And dielectrics are dielectrics – they can only hold off so much electromotive force before bleeding. Raw physics and properties of materials.
Hence – in a public utility environment, to store energy surges would require ginormous (that’s a technical term) capacitor fields. As far as the eye could see. Simply not economic.
Just sayin’
I could be wrong, not my engineering field of expertise.
You might want to take a look at the Fairbanks array. About ten years ago Fairbanks, Alaska installed a BIG array of batteries as part of its power system. The Fairbanks array can provide enough electricity for about 12,000 residents for seven minutes. It’s the size of a football field. Doing the same thing electrostatically would require a lot larger.
Also there’s a Texas company called EEStor you might want to take a look at. One of the problems in using capacitors is current leakage at high voltages.