"The more I find out, the less I know."

Sun - July 8, 2007 at 08:46 PM in

The Solar Equation

On the heels of my article yesterday, I decided to crunch a few numbers about how much it really would cost to switch to photovoltaic (solar electric) power for home use.

I did some research, found a few numbers, and crunched them in a spreadsheet detailing how much per year it would cost to power using solar power.

A lot of people think in terms of how many years it will take to pay back the cost of the system, but the more relevant question is whether the system will cut your electric bill enough to pay the interest on a loan to fund the cost of building it. Any savings in excess of the interest expense can pay down the principal of the loan and eventually go into your pocket; or, if the system can't cover the interest cost, the difference has to come out of your pocket.

The relevant numbers are the cost per watt (installed) of the solar system, the energy generated per watt of the solar system over the course of a year, the retail price of electricity, and the long term interest rate (i.e. mortgage rate).

The range of estimates for the cost of a solar system is $7.50 to $11/watt, with larger systems being less expensive per watt. In Minnesota, one watt of solar panels will generate between 1.0 and 1.1 kilowatt-hours of energy per year on average (assuming that you've chosen a good site), which actually compares favorably with Texas and Florida.

On our last power bill, after all the fuel surcharges and whatnot, Excel Energy charged about 9.3 cents per kilowatt hour. Minnesota's net metering law says that the utility has to buy power from small producers (i.e. residential solar systems) at the retail price, even if the net usage is negative. That makes the calculation simpler as compared to some other states, where the utility is only required to pay wholesale prices.

For the interest assumption, I used 7%, which is the ballpark for long-term fixed interest mortgage loans these days.

The bottom line of all this is that with realistic assumptions for today's economy, each watt of installed photovoltaic capacity would cost about 43 cents/year: 52 cents to finance the system, minus 9 cents of power sold back to the power company.

Another way of looking at this is that photovoltaic power is currently five times as expensive as the stuff Xcel provides, and someone who installs a solar electric system to supplement grid power is subsidizing 80% of the cost of the electricity being sold to the power company.

This isn't the end of the story, though. Right now, Minnesota has a $2/watt rebate program for grid-connected solar installations, which brings the net cost per watt down to about $0.30 per year; or you can think of it as the power company is paying 20% of the cost of the system (from selling power back to the grid), the state is paying 20% of the cost, and the customer is subsidizing the remaining 60%.

Unfortunately, this rebate program is limited to $1 million total, and from what I can tell it may already be fully subscribed--meaning that no rebate is available for systems which aren't already being built. There's also a federal rebate which will pay 1/3 of the cost of a new photovoltaic system, but that's limited to $2,000 total, making it relatively unhelpful for larger installations.

Looking Ahead
Fortunately photovoltaic technology is still advancing rapidly and we can expect the costs to drop fairly rapidly in the coming years. Currently about 75% of the price (installed) seems to be for the parts, and 25% for the labor of installation. Of the parts, about $3-$5/watt is for the solar panels themselves, with the remainder (or about a third of the components) for batteries, inverters, circuit breakers, etc.

Right now the price of solar modules is relatively high because of a global shortage of silicon wafers, so it's not unreasonable to expect prices to drop by half within 3-5 years as new capacity comes on line. Other components are also likely to get less expensive as production volume increases and technology improves.

I would also expect installation prices to drop as companies gain experience, techniques improve, and new systems are developed which are simpler to install.

Is it reasonable to expect $5/watt installed photovoltaic systems within five years? I think so, especially for larger systems. It's also possible that we could see $0.15/kWh for electricity in the same timeframe if fuel prices keep going up. Those two assumptions cut the effective price of the photovoltaic system in half, down to $0.20/watt/year.

Longer term, it's not hard to see where the economics could actually come out positive:

1) A switch to market-based metering, where the price of electricity varies according to the actual demand for electricity at a given time of day. This favors solar power, since the peak power output matches neatly to the highest peaks of electric demand (sunny days in the middle of the summer). Given that $0.093/kWh is the average price (round the clock), it isn't hard to imagine that a photovoltaic system could yield over $0.30/kWh with a market-based price.

1a) Even in the absence of market-based pricing for electricity, it's not unreasonable to expect $0.30/kWh within 15 years if fuel prices continue to increase.

2) A drop in installed cost to around $3/Watt, driven by a 75% drop in the price of solar modules (not too hard to envision), plus a significant drop in the cost per Watt of installation and other components.

These two assumptions (and the 7% interest rate) give you a system which generates 9 cents/year in excess income from selling power, after paying the interest expense for the solar array. Assuming a lower interest rate or slightly better site selection (yielding 10% more energy per year per installed watt) rapidly increase the profit margin.

So the bottom line is this: With today's prices, you can expect to pay about five times as much for solar power as from the electric company, at least here in Minnesota. This is ignoring the environmental benefits of solar power, of course.

But with some reasonable assumptions, it's easy to see how this might be only 2-3 times as much within five years, and it's possible--given increasing fuel prices, market-based metering, and dropping photovoltaic prices--that solar power could pay for itself within 10-15 years.

Posted at 08:46 PM | Permalink | | |

Powered By iBlog, Comments By HaloScan
Syndicate this site