Editor's Note: Scientific American's George Musser will be chronicling his experiences installing solar panels in 60-Second Solar. Read his introduction here and see all posts here.

The first step in installing solar panels, not surprisingly, is to call an installer. I contacted one through Home Depot and, on a Saturday earlier this month, Cameron Christensen of 1st Light Energy came by to give us a quote.

Here's what I learned: The size of a subsidized system is limited by the size of your roof—more precisely, the portion of your roof that has the right orientation and illumination—or by your energy consumption, whichever is less. In other words, you may end up with less capacity than you need, but never more.

In my case, 24 panels, each spanning about 1 by 2 yards and producing 200 watts of electricity, will fit. My college friend Greg Feldberg has fit on about a third that many and tells me he has panel envy (but he has a swimming pool, so we’re even now). All of the panels would be on the back side of the house, away from the street, which is important as we live in a historic district. They’d point due south and have a tilt of about 10 degrees, which Christensen described as a compromise between squeezing as many panels as possible on the roof and maximizing their power output at our latitude in suburban New Jersey.

Unlike a solar hot water heater, which requires direct sunlight because you need to heat water to some minimum temperature for it to be useful, photovoltaic panels continue to produce some power even when it’s cloudy. For a seasonal average of 4.3 hours of sunlight per day -- you can calculate your own situation here -- and factoring in DC-to-AC power conversion losses, that means a total of 14 kW-hr per day, which is exactly what we currently consume. So the system would zero out our electric bill, which averages about $80 per month. (Unfortunately, we’ll still have to pay for gas heat, although I’ve also been working hard on insulating and weatherstripping—more on that in a future blog entry.)

The total cost would be just over $40,000. That's for a full system, including the panels, power inverter, new electric meter, and plenty of blinking lights to show off. The installer handles all the permits and legal whatnot. The system does not include batteries for power storage. Whenever the panels produce more electricity than the houses uses, the grid absorbs the excess and we draw it back later, a concept known as net metering. How exactly the utility manages net metering is a technical challenge in its own right. I'll tackle that in a future post.

Now we get into the tangled subsidy structure. The state ponies up $1.75 a watt, for $8,400. (There’s some disagreement over whether this is taxable.) The Federal credit kicks in 30 percent of the cost after the state subsidy, for about $10,000.

The tricky thing about this credit is that it is not refundable, which means that, at most, the tax credit will zero out your IRS bill. In other words, the credit will pay 30 percent of your solar system cost, or your full tax bill, whichever is less. So you can take full advantage of it only if your income tax bill exceeds $10,000. This makes the subsidy somewhat regressive. Be that as it may, I’m lucky enough to have paid at least this much tax. (That is the first time I think I have put the words “lucky” and “tax” into the same sentence.)

Our utility, Public Service Electric & Gas, offers an attractive loan program based on SRECs—solar renewable energy certificates. SRECs are sort of a baby step to a full carbon-trading system. They are used by states to meet their goals for generating a certain fraction of their electricity from renewable sources. A single SREC represents the generation of 1 MW-hr of energy and you sell it to help recoup the cost of the system. For the consumer, the trouble is that the price floats up and down depending on market conditions. The utility loan program eliminates this risk by putting a floor of $475 under the price. For this privilege, though, you have to pay the loan interest of 6.5 percent, and if the SREC price floats higher, the utility keeps 25 percent of the difference.

In my case, my panels would generate about five SRECs per year, for a total of about $24,000 over 10 years at the floor price. The utility will front this much money, less interest, for purchase of the system. So that comes to about $17,000.

The bottom line: For the $40,000 system, I need to come up with $5,500 out of pocket. I can put that amount on a Home Depot credit card at zero interest and later shift it to a home equity line and deduct the interest on my taxes. Because I now spend about $1,000 a year in electricity, that means the system pays itself off in about six years. It could be faster or slower depending on how the system performs, how the taxes shake out, and what the SREC price does.

These subsidies are what makes solar possible these days. If we had to pay the full cost, it might, in principle, still be worth it, depending on how long we intended to stay in our house and how much it added to the value of our home. The panels are supposed to last several decades. But the average person just can’t come up with 40 grand, no matter how wise an investment it might be over the long haul.

The back part of our roof, photo courtesy of Cameron Christensen