My Experiences with Solar Electric and Solar Hot Water Systems
An energy management consultant since 1974, Andrew Rudin shares his personal accounts, the good and the bad, with using solar panels to generate electricity and to heat water.
I have been an energy management consultant since 1974. Before suggesting that someone buy an energy-saving gadget, I evaluate that gadget myself. In this article, I describe some of my personal energy-saving gadget experiences.
My Solar-PV System
In 1998, I got interested in solar PV. Earlier that year, our local combined electric and gas utility (PECO Energy) changed its electric rate to allow net metering for solar-electric systems. Net metering eliminates the need for batteries, which increase system maintenance, generate combustible hydrogen, and decrease system electric production. Electrons move into the grid or into our home/office, depending on the amount of available sunlight and our need for electricity. My 2.7kW system started operating in July 1998. It was pictured on the cover of Home Energy magazine's Sept/Oct '02 issue.
The system cost $26,180 -- about the same as a new car. I received two turnkey bids from contractors, but I thought the prices were too high. I hired a solar system designer -- Ron Celentano—and ordered the system components from California suppliers. Hundreds, if not thousands, of PV systems had been installed in California. I hired a roofing contractor to install the panels and a commercial electrician to wire them. At that time, there were no rebates. I took advantage of the federal 10% tax credit. Since the system is installed on a home/office, it also qualifies for accelerated depreciation, but my accountant at the time didn't know that. So I lost my first (and greatest) tax advantage that first year. I replaced that accountant with one who brought in a check from the IRS for $3,645, for accelerated depreciation in the second year. That, plus a couple more years of accelerated depreciation, reduced the initial cost to about $19,000. The only maintenance cost on the system so far has been $100 to find and replace a blown fuse, which is amazingly low for such an expensive investment over a dozen years.
The electric metering is complicated. In late February 2002, the utility replaced a single net meter (one with a disk that moves one way or the other, depending on how much the sun is shining and what my electric loads are) with two electronic meters, with digital displays on the rear outside wall of my house. The In meter measures the electricity that comes into the home/office, and the Out meter measures the electricity that goes out. Since the Out meter is wired in reverse, both meters display positive values as the power passes through either one or the other. Since I am interested in net metering, I installed an additional conventional disk-spinning meter, so I could observe which way the power was flowing (the disk spins backward or forward). To visitors who may not understand solar-generated electricity, the backward spin of the disk inside this meter is the most visible (and often remarkable) evidence of what the system does.
From the beginning, I have had a fourth electric meter that measures all of the kWh produced by the PV system. This fourth electric meter is the solar meter. In June 2001, I replaced the standard kWh solar meter with one that also reads maximum kW power production. In the basement, I also have a standard electric meter that measures all the electricity that the home/office uses, regardless of whether it comesfrom the grid or from the solar-PV system. I read these five meters every day; on the 15th and last day of each month, I record the data on a spreadsheet. I record the solar peak kW by resetting the solar meter and subtracting the reading, estimated to 0.01 kW (10 watts), from the reading that I took the last time I reset the meter. When I am away on vacation, I read the meters when I return and prorate the kWh for each day. The peak kW would be the peak for all of the vacation time.
On the roof, there are 36 75W British Petroleum (BP) panels. When the sun hits them, they generate direct current (DC) electricity that is collected by wires that travel through the roof to an inverter in the attic. The inverter does three things. First, it collects all the DC electricity and converts it to the alternating current (AC) that the home/office uses. Second, it receives the frequency of the AC power input from the utility and synchronizes its AC output to correspond. And third, it increases the voltage of the solar-generated electricity by a few volts above the utility voltage, which is enough to move electrons back into the grid, if the electric load is satisfied in the home/office. When sunlight is barely sufficient the inverter makes a quiet thumping sound, which stops when the sun is either too dim or sufficiently bright for the inverter to generate alternating current. Unless I am hearing that sound, or watching the electric meters, I can't tell that the system is working.
Because I receive solar renewable energy credits (SRECs) from my utility, I make about 13¢ for each kWh that my system generates, whether I use it in the home/office or not. Since May 2005, I have sold my SRECs to the Energy Cooperative of Pennsylvania for about $2,400. (In 2009, for example, our net cost of electricity and gas was $385 for the year.) As part of the deal, I buy 100% renewable energy from the cooperative.
Table 1 shows our annual kWh use and PV production for the years 1999 through 2010. The average of the kW readings is the average of the bimonthly peaks for the year, with a few less than 24 bimonthly meter readings to account for summer vacations. The average peak numbers show that the base kW is not that different from the average kW. I found this interesting because it shows the peak production for the year is not that different from the average of the bimonthly readings for the whole year. The peak for the year is the maximum of the readings. Table 1 does not give figures for 1998, because that was a partial year. There were no kW recordings until I installed the demand meter in June 2000. The ups and downs in production are likely due to changes in solar irradiation, atmospheric clarity, and cloud cover. The overall trend since I installed the system, however, has been a 33% reduction in generated electricity and a 17% reduction in peak generated kW.
The columns on the right are the number of days each year that the panels did not generate 1 kWh and the percent change from the previous year. Note that this number increased by more than 550% as the panels aged.
The electricity we use in our home/office has been consistently low. The average home in PECO Energy's service territory uses about 7,200 kWh per year, so at 2,289 kWh, we use less than 1/3 of the average. Not bad for two adults, five computers, a home entertainment center, and all the usual appliances. We use natural gas for space heating, cooking, clothes drying, and heating domestic hot water (DHW).
Not So Simple Payback
I look at payback in several ways. Simple system costs payback is a very weak reward for me, given the extensive publicity the systems have provided—publicity that, I hope, has furthered the cause of renewable energy. I installed the solar systems as part of an effort to improve my lifestyle. I gave away my car at about the same time as I installed the systems, saving me about $6,000 per year. All this happened after I talked with some Amish farmers in Lancaster, Pennsylvania. As you probably know, they refuse to be connected to a utility grid, or to own or drive a car. They do this, not because they are looking for payback, but because they believe that to live simply improves the quality of life for their families and communities.
Since the utility installed its two new unidirectional meters in February 2002, they have measured 14,144 kWh coming into the home/office and 12,466 kWh going out, over about nine years. So let's assume that the home/office requires about 190 kWh per year more than the solar-electric system produces. The current price of electricity, based on my October 2010 electric bill, is $0.216/kWh. This includes electric distribution, transmission, and transition, and the premium renewable-energy supply. Essentially, I have a feed-in tariff agreement, since I pay my renewable-energy supplier for all of my electric usage. Each year, then, I pay about $41 per year at current prices.
The Energy Cooperative of Pennsylvania has paid me about $2,400 since June 2002 (6 1/2 years) for about 17,500 generated kWh (the SREC I get from PECO). At 13¢/kWh, my generation pays about $350 per year. Subtracting the $41 I pay PECO, my net annual gain is about $310 per year. So my simple payback is $19,000 divided by $310, or 61 years. I am not sure what the life of the array will be. BP has backed off its warranty, so I doubt it would replace a bad module even today. It is possible that the system could keep working for many, many years—just at much lower production than I was getting to begin with (see "Not So Simple Payback").
In addition to lowering electric costs, the system also paid off in other ways. About three years after the PV system went on line, the Philadelphia Inquirer asked me for an interview. They were doing a story about three independent power generators, and I was one of them. The other two, however, didn't want to be included in the article. So the front page of Philadelphia's main newspaper (circulation about 380,000 copies at the time) on Wednesday, May 9, 2001, had a picture of me, with the PV panels in the background, and a very flattering article.
I was a local hero. A few days later, I received a call from NBC's Nightly News with Tom Brokaw. After two days of videotaping, I was featured in a 41-second segment about my system on May 17, 2001, just after a segment in which George W. Bush announced his national energy plan. At the time, a 30-second ad on this program would have cost about $55,000, according to a local ad agency. So my segment was worth about $75,000, if I had bought the airtime, never mind the production costs. The impact of that segment on my reputation was huge. I was now a national hero.
Around 2000, Pennsylvania began a solar-PV grant program with funds originating from a rate case settlement from PECO Energy. Although it had little to do with my PV system, they hired Ron Celentano, the guy who designed my system, to run the program, and he hired me to help, working part-time. From April 2004 through January 2009, I earned just over $20,000 in consulting fees. Without a doubt, my system paid for itself -- and not only with money saved on electricity.
What about the pure economics of the system? Was it worth the investment purely from an energy point of view? The late ecologist Howard Odum was interested in what he called the emergy of a PV system—its embodied energy. I sent him all the data he requested about the copper, aluminum, silicon, and plastic in my system. He determined that the electric production was borderline worth the energy it took to manufacture the panels, inverter, wire, and so on. That was in 2002. The system is producing less energy now, so it might not meet Odum's criteria today.
So while everyone has a roof, I have one that pays me back, albeit very slowly to date. The inverter seems to be the most crucial part of the system. If it breaks, I will have to pay thousands of dollars for a new one.
I was optimistic when I began this project 12 years ago that solar energy would achieve wide acceptance. But I then heard the coordinator of the solar-PV grant program describe solar electricity as "Gucci power." I felt terrible about that. More important, all I hear and read about solar PV does not acknowledge some basic truths that I have learned.
When they are covered with snow, the panels don't work. We carry on because our system is net metered, and electricity is not interrupted. This means, however, that while we displace PECO's kWh with sun-derived kWh, PECO has to maintain sufficient generation capacity to serve all of the now more than 1,000 PV systems in its territory. The kWh the PV systems do displace are likely to be produced on sunny days in the summer. This may offset the high cost of operating some of PECO's natural-gas peaking generators.
PV systems have improved. They operate more efficiently than my system did, even when it was first installed. And the annual degradation of the solar cells isn't as high for PV modules manufactured today as it is for the older ones that I have. My system cost about $9.70 per watt to install (excluding tax benefits), but now, according to Ron Celentano, the median installation cost for a residential system is about $6.10 per watt, and could be under $5 per watt excluding rebates and tax benefits. The federal tax credit for PV is now 30% of installation costs compared the 10% that I received 12 years ago. And I could have gotten as much as 36¢/kWh for my SREC for part of the life of my system, compared to the 13¢/kWh that I currently receive. Finally, electric rates are higher today than they were 12 years ago, and will probably increase as time goes on.
I don't know if the system increases the resale value of this home. My guess is that it doesn't, although EPA says that it should.
My Solar Hot-Water System
I use our local library a lot, saving money by borrowing instead of purchasing books. One librarian asked me to look at the library's electric and gas consumption, and suggest how the library might reduce energy costs. When I examined the building, I discovered an abandoned solar hot-water system on the roof, probably installed when the library was built in 1982.
Eventually, the library wanted to replace the roof and the rooftop HVAC equipment. The panels were in the way, so I asked for two panels in trade for my energy-consulting services. I installed them on the roof of my sunroom, below the main roof on which the solar-electric system is installed. In December 1998, I hired a contractor to do the installation. Cost to install the system was $2,450, not including the two free panels. The 10% federal tax credit was $245.
The system uses a 65-gallon storage tank to preheat water. When there is adequate sunlight, a small (10W) PV panel generates enough DC current to heat a 3-inch cylinder of liquid. This liquid expands to push a brass shaft to open a valve. The open valve allows pressurized water from the local water utility to fill the panels until an air vent at the top of the system closes.
The shaft continues to open until it presses a switch that operates a variable-volume DC pump. The pump circulates water through the storage tank and the two solar panels, where the water is heated. When there is insufficient sun, the process reverses. As the shaft retracts, the pump turns off. Then the valve opens up to drain about 2 gallons of water either into an interior drain during the winter, or into the garden at other times of the year. A vacuum breaker allows air into the panels. Late one fall, I forgot to pull the drain tube out of the garden. Water in the tube froze, preventing the drain-down. The next morning, water was blowing off the roof into our garden. A copper pipe had burst, costing me $100 for a plumber to replace the broken section. I could have lost the whole system that night.
The solar system preheats the water from the local water utility in the 65-gallon tank on its way to our backup DHW system. Originally, the backup system was a stand-alone gas water heater vented through the wall. After the experiment described below, I replaced it with an indirect-fired tank that is heated by our high-efficiency natural-gas water heater.
To estimate the savings from the solar DHW system, I bought and installed a natural-gas submeter, just to measure gas for our vented domestic water heater. For the year ending in May 1999, the submeter measured 99 hundred cubic feet (ccf) of gas. Then I turned the solar system off but kept reading the meter. For the next 12 months, the water heater used 145 ccf without the solar system. Then I turned the solar system on. For the next 12 months, the water heater used 86 ccf with the solar system operating. My wife and I were away on vacation for about the same number of days during each of the three summers.
The solar DHW system saved me 52.5 ccf per year. The October 2010 PECO invoice shows a gas commodity price (excluding the monthly meter charge) of $1.03/ccf, so the 52.5 ccf has a current value of $54.
Gradually, the air vent/vacuum breaker stopped working, allowing water to slowly spray out. Then, in the spring of 2008, the variable-speed DC pump stopped working. I decided to replace these things myself, because the materials alone cost a steep $344.50. The system has been working flawlessly ever since.
Andrew Rudin can be reached at firstname.lastname@example.org.
So, the total cost of owning this solar DHW system has been about $2,650, including the repairs but excluding the cost of the free solar panels. If the system had operated continuously since January 1999, I would have saved the current value of the system, which is about $600. The payback on my system is about 49 years, if nothing else breaks down, which is unlikely.
Being an Early Adopter
I showed this article to Ron Celentano. He e-mailed back: "You don't give yourself enough credit for being a pioneer with this technology -- or at least you don't convey that in your article. Unfortunately, it's the pioneers that get burned the worst, but they also have the first thrill of surfing the wave, before the boring mainstream sets in."
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)
While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.