Moore’s not here

Chart - Price of Chinese C-Si modules
Price in C$/watt. Data courtesy Solarserver.

Late last summer, I was walking the neighbourhood near where a rooftop residential solar PV system was about to be installed. People tend to be more interested and less sceptical about solar when their neighbours are getting into it, so it’s good policy to pound the pavement in the hopes of drumming up some business. I had already had several great conversations, and had some promising leads jotted down on my smartphone.
I saw a gentleman sitting on his porch, engrossed in the notebook computer on his lap. His roof was a good candidate, so I introduced myself. He obviously wanted to escape the conversation as soon as possible and get back to his surfing, so I did my best to keep it brief. I explained that his home had a good roof for solar, that some folks down the street were about to get a system, and there was good money to be made in generating electricity from sunshine.

The man clearly believed he knew all he needed to know about solar, and even more clearly demonstrated that he knew next to nothing about it. He explained that he worked in the information technology field, and IT equipment is typically amortized over three years. He had no intention of making a 20-year technology commitment when whatever he installed would doubtless be obsolete before, say, a politician could complete one term in office.

I won awards on my high school debating team, so I can make a persuasive argument. I’ve trained people in the corporate world on things like email encryption tools and document management systems, so I can explain things. I’ve also had to make sales pitches to C-suite executives, so I can convince people that what they’ve just learned from me is, in fact, simply something they knew all along.

However, I’m also a reasonably good judge of character. I could tell that with this guy, I’d be wasting my time. That he knew IT cold, I have no doubt. But he greatly overestimated the similarities between his field and others. Microchips and solar cells both depend on silicon, but concrete does too, and the three have about as much common ground. I wasn’t going to convince him that he was missing out on a great opportunity. Better to mosey on over to the next likely house and hope for someone with an open mind.

Nonetheless, the fellow was wrong. Dead wrong. Moore’s Law does not apply in the world of solar photovoltaics. If you look at cell efficiencies, nothing terribly exciting is going on. Certainly nothing like a doubling every 18 months, which is something that has become an article of faith in IT circles.

Oh yes, every time you turn around there’s another breathless announcement by some university research team that has made an astounding breakthrough in efficiency. However, these innovations seem to disappear on the road from the lab to the marketplace.

There are plenty of reasons. The devil is in the details, so maybe the process to manufacture the new-fangled device is too complicated, and cannot scale to mass production. Solar panels have to be able to take a lot of abuse, so maybe the end product isn’t robust enough to withstand shipping, installation, and two decades of weathering in every earthly climatic zone. Efficiency is useless if it comes at too high a price, so maybe, as in the case of solar cells known by the acronym CIGS (Copper Indium Gallium Selenide), three of the four key components are so rare that manufacturing large volumes with that stuff would use up the entire global supply and send the price into the stratosphere.

If you look at mainstream solar PV installations, most are the breed known as C-Si (crystalline silicon). The processes for transforming silicon into ingots, ingots into cells, and cells into modules are well established, simple, and efficient. The finished product is solid and easily able to put up with 20 years of whatever Mother Nature can throw at it (while still remaining fully functional, and retaining 80-90% of its capacity at the end of that time). The materials aren’t terribly exotic. The principal element happens to be the second most common in the earth’s crust after oxygen. On every Caribbean beach, you see tons of it – sand. We won’t be running out of that any time soon.

This time last year, most of the panels being sold were 245 watts. This year, most are 250 watts, with some 255-watt units thrown in. A 4% increase in efficiency over an entire year is not going to impress anyone, especially my IT-schooled front porch friend. You can rest easy knowing that if you install a solar array on your roof this year, the ones being installed the following year won’t be dramatically different.

That doesn’t mean that the industry is static – far from it. However, the big news is not gains in efficiency; it is reductions in price. These come from scale economies, meaning that the more units you produce, the cheaper each unit becomes. It’s not as sexy as press releases touting scientific breakthroughs, but it’s here and now rather than pie in the sky.

Let’s take Chinese C-Si modules as an example. If you look at Solarserver data from September 2011 to January 2013, the price per watt in any given month is between 50% and 65% of what it was a year earlier. At no time during that time period did prices rise. Cell efficiency did not play a significant role in that downward pricing trend. It’s all about economies of scale.

In personal computers, it’s easy to make the argument that you should wait until next year’s model, because it will be cheaper and more powerful. You could make the same argument in solar PV. Why buy this year, when next year the cost of panels (and, as it turns out, the entire system) will be lower? The problem with that rationale is that you never end up buying a computer, or a solar PV system.

The real question you need to ask is whether it makes economic sense to buy today, based on prices today, and based on benefits today. The fact that a newer, shinier, more powerful product comes out next year has no impact on the validity of that decision.

Even for Mr. Porch Surfer, tomorrow never comes.

2 thoughts on “Moore’s not here”

  1. Nice article,
    I’ve been interested in solar for 30 years, and have done a few experiments with solar concentration. Of course electricity from solar is one of the most practical solar energy applications. I was stunned when I did a recent search on solar cell prices to see how low they have fallen. That’s the good news… The bad news here in the states is cost of installation. There are local companies in my area (St Louis, MO) That have been advertising solar installation, and it’s clear from the prices they are quoting that you will probably be paying 3 times the cost of the panels for the installation. An ad they run starts by announcing that a typical residential system would cost $40,000 US, but with Fed, State, Local rebates you can get that system for $12,000. So right off I feel like I’m being ripped off because I know the cost of those cells is now probably under $1 a watt. So to make solar more attractive at this point you need to make it more do it yourself friendly, or bring down the installation cost curve. I work for Baldor Electric, a major Electric motor MFG, and I find myself wondering why they don’t package and sell an off grid solar panel- motor package. It would be ideal for pumping water in remote locations. 4 typical solar panel array would easily power a 1 Hp motor.
    Here’s another great Idea, Pair up a small solar installation dedicated to work with your Home’s central air conditioning system. That way you would always be getting the most power exactly when you needed it. While you didn’t mention it, you probably know that “with every doubling of solar mfg capacity there is a corresponding reduction in price” of course This can’t go on forever, and don’t expect the price of solar panels to drop much below $.50 a watt, but who knows? And if my math is correct at $.50 a watt Versus $.o1 a KW purchased price it would take 5,000 hours to pay back the actual cost of the cells. 5 hours a day of solar average per day times 365= 1,825 hours per year, that’s less than 3 years payback. not bad. But once again you have to bring down the total system cost. Inverters, wiring, installation, and solar panels. The total installed price needs to come down to less than $2 a watt, then you’ll start getting peoples attention.


    1. Thanks for your comment, Levi. You are correct that the panels only account for a fraction of the cost, and with the panel price dropping, that share of the cost is decreasing. There’s also racking, inverters, cabling, meter installation, utility account setup, engineering structural review, building permitting and inspection, electrical safety inspection, and other costs that have nothing to do with the panels. Installers have little control over most of these costs, and many will apply even if you have the skills and courage for the DIY option. If you know the cost of framing lumber, sheathing, siding, insulation, and all the other materials that go into building a house, you may also feel ripped off by the final price tag – but there is value in the service that the builder provides, and they are in the business to make a living and not to provide charity.
      Some of the costs are fixed, so coming up with a price per watt for a completely installed PV system is not possible; these costs would still apply regardless of the size of the system. If the fixed costs amount to $5K, your $40K system (likely about 10kW nameplate capacity) runs $4 a watt, but your neighbour’s 12kW system will come in at $3.91/watt. The tipping point in the price also depends on where you are and how much electricity you consume. If consumption is high, electricity is expensive, and the solar resource is plentiful, solar is already the best way to go. In other words, the $2/watt figure may work for you, in the state that ranks 47th for electricity prices. I calculate that the figure here in Ontario is $2.31/watt for an average residence with average consumption – but that won’t be the case for someone in a different geographical area. Solar is already the cheapest energy option for tropical islands like Hawaii, where the sun is bright and the incumbent energy source is expensive imported diesel. (By the way, the purchase price of electricity is usually stated in $/kWh, so I’m not sure how to interpret your $.01/kW figure.)
      As for your “killer app” suggestions, there are already such systems available. For example, a company near my location sells packaged “mini split” ductless air source heat pumps powered by solar PV panels with grid backup, to provide both heating and cooling. I really like your off-grid panel/motor pairing idea – it’s a perfect solution for solar pool heating, since the very time you want to run the pump up to the solar collector is when the price of grid electricity is highest. Talk to ! I’m sure there’s money to be made there.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: