Going Solar

The Luce Abode

  • Motivation
  • Looking for Contractors
  • Site Survey
  • Proposals-Cost & Savings
  • Comparison of the systems
  • Financing
  • Decision
  • Bureaucracy
  • Redesign
  • Installation
  • Official Production
  • The First Month
  • The First Year
  • The Second Year (TOU)
  • The Third and Fourth Years (Games with Baselines and Surcharges)
  • The Fifth and Sixth Years
  • Photos
  • Lots of Photos
  • Data, Charts, Graphs
  • PV Payback Period Estimate
  • FAQ
  • Solar Attic Fan
  • 2006 Solar Home Tour
  • 2009 PV is Viable in Santa Clara!
  • Links
  • Motivation

    Shortly after we bought our house in 1999 in Redwood City, where the city motto is Climate Best By Government Test, we thought about adding solar photovoltaic (PV) panels to our roof. Afterall, this was California where the sun was plentiful and the electricity rates high. Also, it's good for the environment.

    We'd read about other people who had done it. (Guerrilla Solar) The cost of a system would be $20K or more, and having just bought the house, we couldn't afford it. Still, we thought we might eventually do it.

    Then came 2001, the year of the California Energy Crisis; a crisis quite possibly contrived by the energy companies. So in July, our electricity rates went up again in a tiered rate system, where power used over 300% of the baseline cost $0.245/kWH. Over half our electricity usage was in this top tier. With California offering $4.50/watt rebate on a solar PV system, and PG&E being required to accept hookups (because of the net metering law), solar was sounding even better. Still, a system would take roughly 14 years to pay for itself in reduced power bills, or so I thought.

    I met someone who pointed out that with the California rebates and the record low mortgage rates and the mortgage interest tax deduction, you could save more on your electric bill per month than you'd have to pay on the mortgage. By my initial calculations, this wasn't exactly true, though maybe it would have been with the federal deduction you get if you have a home office. But it was true if I added in the monthly savings from refinancing our original home mortgage at the current lower interest rates. I was sold.

    The rolling blackouts of 2001 were caused in part by the lousy electrical distribution system in CA. There was energy to be had, but it was difficult to get enough of it to some places that needed it the most. By generating electricity and putting it into the grid during the peak usage hours, we will do our part in reducing rolling blackouts in our area.

    We also thought we'd get a natural gas powered backup generator installed at the same time as part of the whole system. The last 17 hour blackout we had, all our friends were asking "what happened to apricot.com?"

    We are also planning to add a hot tub inside our gazebo. Not wanting to drive up our electricity costs a lot with heating the water, we also wanted to get a solar hot water heater to mount on the roof of the gazebo.

    Looking for Contractors

    How to find a Contractor? I searched around the web and found the California Solar Energy Industries Association list of member contractors. Considering only the ones in the SF Bay Area, I went to the California State License Board site and checked to see that they were licensed, bonded, and had workers compensation. Next check was the Better Business Bureau site to see if they were members and what their records were.

    Still, how to narrow things down a bit more? One recommendation was to look for contractors that are licensed for electrical work, C10 and Solar C46. That, and my preference for places that have websites, narrowed the list to three. I called up SolarCraft Services and found they only do work in Marin and Sonoma counties. Sun Power & Geothermal Energy (now called SPG Solar ) and Akeena Solar each came to our house to discuss what we were looking for, and to do a site survey.

    Curiously, despite having a C10 electrical license, Sun Power did not do backup generator installations, though they could subcontract it out.

    Site Survey

    The contractors looked at our electricity bills for the past year and entered our KWH usage by month into their spreadsheets. They wanted to know our goal. Was it to eliminate our electric bill entirely, or to get the most return on the investment? We didn't think we had enough roof space to do the first, so it was either the second, or a third option, to cover our roof with as many panels as would fit. Our house is 1170sq.ft. not including the garage. Each kW of panels takes up roughly 100sq.ft.

    They measured our roof, including the direction and pitch of each face. The south facing surfaces are the best, and a 30 degree pitch is optimal. Our roof has a 22 degree pitch, and our house faces slightly south of east. Both companies seemed to think a system of about 5kW would give the best return on investment, but that would not all fit on the SSW facing roofs. Akeena felt the remaining panels should go on the front facing roof since, by facing slightly south, it would get more light than the back roof. Sun Power, on the other hand, thought changing from a PG&E E-8 (old E-8) seasonal schedule to an E-7 (old E-7) time-of-use schedule would save us the most money, and recommended putting the remaining panels on the back facing roof, hidden from the street.

    On E7, power used during the peak usage hours of Monday through Friday, noon through 6 p.m., is charged at the rate of $.32524/KWH (including the 1/4/2001 energy surcharge) in the summer, while off-peak is closer to $.095/kWH. The thought was, with West facing panels, we could run the meter backwards in summer during the peak hours and get credited $.325/kwH by PG&E. However, after redoing the calculations, Sun Power said we'd only save $100/year on the E7 schedule vs. the E8 schedule. The E7 schedule requires a special time-of-use meter be installed.

    Still, with PG&E asking for more rate hikes, we wanted to consider a system larger than 5kW. Akeena, which uses 2.5kW SMA inverters felt that 7.5kW would be the next logical step up. They would put the additional panels on the back roof with reverse-tilt mounts to face more southward and thus get more power per panel than if the panels were flat against the roof. I didn't mind having the reverse-tilt panels in the back where they would not be visible from the street.

    Sun Power, using both 1.8kW and 2.5kW inverters thought the next step up would be a 6.8kW system, by adding 15 panels to our front roof. I'm asking them if they can fit 22 panels and go up to 7.4kW system. Even at 7.5kW, most of the power we'd be replacing would be at the highest tier surcharge rate and a little would be at the second highest rate.

    Both companies said they would handle the building permits and the PG&E and city inspections. Installation would take about a week, inspection would be a week or two after that, and the $4/watt rebate would arrive a month or two after that. They would reserve the rebate for us, and would not start installation until the rebate is approved. The rebate was $4.50/watt through October 2002 and then dropped down to $4/watt. It was January, 2003. The rebate had not yet started up for 2003, but was expected to in February. The inverters have a 5-year manufacturer's warranty and the PV panels a 25-year manufacturer's warranty. In addition the contractors give a 5-year warranty on the entire system.

    Proposals-Cost & Savings

    Sun Power offered us a 5.2kW system at $51,475, or $28,074 after the $4 per watt California rebate(broken link?) and the 15% California tax credit. They estimate this will save us 9,000kWH or $2,100 per year, reducing our annual electric bill from $5,400 to $3,400. A 6.8kW system would cost $67,987, or $35,062 after rebate and tax credit, cutting our annual electric bill roughly in half by $2,700 or 11,700kWH. At my request, they later quoted a 7.4kW system. The numbers were $73,265, or $37,543 after rebate and tax credit with an estimated 13,700kWH or $3,100 savings.

    Akeena offered a 5kW system at $52,103, or $27,472 after rebate and tax credit or a 7.5kW system at $76,911 or $40,278 after rebate and tax credit, a little more expensive than Sun Power. Curiously, Akeena estimated the savings to be 7,600kWH or $1,800/yr on the 5kW system and 11,400kWH or $2,600/yr on the 7.5kW system, considerably less than Sun Power estimated for simlarly sized systems.

    The difference seems to be that Akeena used a more conservative 5.3 solar hours per day, while Sun Power used closer to 5.5. Also Sun Power seems to have rounded our roof angle from 22 degrees to the nearest one on the chart in the more favorable direction.

    All savings calculations are specific to our high electric usage, our usage pattern (pretty flat, no real spikes for heating in winter or A/C in summer) our E8 billing schedule, our location (5.3-5.5 typical solar hours per day), the pitch of our roof and the directions it faces, shade on the roof caused by trees and buildings. This is why the site survey is necessary.

    To get a feel for whether PV is for you, try out the Clean Power Estimator


    We put on a 30 year fixed rate mortgage, (6%, 0 points, $3100 estimated closing costs) the cost of the PV system after rebate. That's roughly $6,000/kW. We put the rebate amount of $4,000/kW on a home equity line of credit (.25% above prime, or currently 4.50%) to be paid off as soon as we received the rebate. This gives us a $36/kW monthly mortage payment.

    By Sun Power's estimates, we'd save $400/kW on our annual energy bill, or on average $33/kW each month. The mortgage interest is tax deductible. Over the life of the loan, you pay an average of $19/kW each month in mortgage interest. At an incremental tax rate of 27% federal, 9.3% CA state, you would save $7/kW each month in taxes. That give a total savings of $33+7-36=$2/kW of installed PV each month.

    It's not much, but it's something. I didn't want to go through the hassle of loan approval and $3100 closing costs for either refinancing our original home loan, or putting up PV, but somehow, putting the two together makes it seem worth it. Our cash flow will improve, but our debt will get larger.

    I have not done a Net Present Value (NPV) analysis, mainly because there are so many assumptions to be made, and a slight change in one makes a huge difference. However, Akeena did a NPV analysis with the assumption of financing the system with a 30 year fixed rate loan at 6.00%, an annual increase in PG&E rates of 5% (It actually has gone up 5.5% on average over the past 40 years), a discount rate of 7% (look that one up yourself. I think it's an assumption that I could have invested that money elsewhere and made a 7% return), and a marginal combined federal/state tax rate of 39%. For the 7.5kW system, they came up with a NPV of $28K over a 30 year life of the system. Also in those 30 years we'd avoid producing 227 tons of CO2 (carbon dioxide, a greenhouse gas).

    I tried to make a more apples-to-apples comparison of the systems. The results were very close.


    In the end we went with Akeena, despite their price being slightly higher. Perhaps it's a reflection on their salesmanship. They were quicker at figuring out just what we were looking for, rather than what people with similar homes might have wanted. They seemed to take into account more properly, the fact that our house was not facing due East. Their layout would more easily allow for expansion, though that may be unfair, since we hadn't told either company about our plans for expanding the house in the mythical future.

    Being more engineering-oriented, Akeena also was able to understand Scanner's request to have the raw data from the inverters go directly by serial port to one of his unix computers where he could generate reports and graphs to study the data, particularly in combination with other data we gather in our house ourselves.

    Akeena also made the process easy. When they presented the proposal, they gave us a binder with details including a project time line with estimates on how long each phase would take and exactly how much money we would pay them at each milestone. They had the Purchase Agreement and other forms relating to the PG&E and the rebate all ready to sign, and in duplicate so we could keep one copy to read after giving them one signed copy. They even included the form for the 15% tax credit.

    We will likely use Sun Power for the solar hot water for the hot tub. However, since it is not likely to have more than a 5-year warranty, I do not want to put it on a 30 year mortgage. There are currently no CA state incentives for solar thermal systems. Sun Power however, wanted to battle for our business


    As soon as the CEC started accepting new applications March 3rd, Akeena turned in ours. We got confirmation on March 26 that the rebate had been reserved in our name. Our loans were approved the first week in March at a rate of 5.75%, which is lower than the 6.00% I was using for calculations. Our contract was contingent on the rebate and the loan approval.

    Then we hit a glitch. Akeena came out and took some more detailed measurements of our roof and took some plans to the city to start the building permit process. The associate planner told them they can not put PV panels on the front of the house or on reverse tilt mounts. Now I've got to find out the exact ordinances and regulations that apply. I've seen PV panels on homes in our town visible from the street, but I'm not absolutely sure any of them are on front facing roofs.

    I found a photo on the web of a house with solar panels on the front roof. The photo was labelled as being of Redwood City. I took the photo to City Hall. The associate planner was not in that day, so I spoke with the zoning administrator who said that you can put panels on the front roof, but they must conform to the pitch of the roof. She also wanted them to look more tidy with fewer gaps between them. She didn't want any reverse-tilt mounts visible from the street.


    Akeena came back to our house to take more measurements on March 25. We decided to try using the new Sharp 185W panels instead of the Sharp 165W panels of the same size. That way we could use fewer panels to get the same power, to avoid having panels hanging off the top corners of the roof.

    We also changed the reverse-tilt mount panels on the back roof to flat mount so we could fit in more panels there, despite losing about 14% collected energy because of the less favorable orientation. It seemed a good tradeoff. That still left 8 panels on the NNE facing roof since we couldn't fit them all some place else. Changing those from reverse-tilt to flat mount costs a 33% loss in collected energy per panel. I am not happy about that and would like to see if we could move some of those panels to a more favorable orientation.

    Remeasurement of the location of the vents on our front roof proved that we had more space than expected on the front, and that we could arrange the panels without gaps between them.

    I re-did the comparison of the systems. The redesign gives us 8% energy less than the original Akeena proposal, and also less energy than the SunPower proposal.

    A few days later Akeena sent me a new roof PV layout. The PV array operates more efficiently if all the panels on each string are facing the same direction. So they put 2 strings of 8 facing front, 1 string facing back, 2 facing SSW and 1 NNE. This layout would give us 7% less energy than the original Akeena proposal.

    I still wasn't happy with one string being on the NNE side, so I sent them a proposal to move the string to the back. They sent me another layout just slightly different from my proposal. This one while giving us 5% less energy than the original Akeena proposal, gives us virtually the same annual $savings. This is because more of the power is generated in the summer when the electricity rates are higher.

    Akeena submitted the building permit application on April 2. The planning department didn't like the look of all the panels on the front, but would accept an orderly 2 x 4 rectangular array of panels in the front. Akeena got them to accept moving the second string of 8 from the front to the NNE side of the house, flush mounted. Akeena sent me the final design on April 9. Since 8 of the panels are once again on the NNE roof, we are back at losing 8% of the energy bill savings compared with the original Akeena proposal.


    The panels and other materials were ordered to arrive on April 25, but rain delays on other customer's projects have pushed the start of installation to May 8, 2pm.

    The first day they brought the mounting rails and started taking measurements on the roof to determine exactly where to attach the rails to the roof rafters. A few brackets were up on the roof by the time they left. They also brought some paperwork for us to read and sign for the $4/watt rebate. Then there was the invoice for the building permit $340.75, which I figure they paid to Redwood City, and the Design Fee $2730.

    The second day they put up some rails on the front and side of the house. They didn't come Monday or Tuesday because of some emergency at their office.

    Wednesday, late morning they came back. They brought the inverters and 16 of the panels on the first trip in their van. Despite the photo in Sharp's brochure, the silicon chips on the panels look black, not dark dark blue. So I was planning to pay the $55,964 charge for the equipment today, but the installation guys disappeared around 3:30pm without a word. During the day, they brought up 8 of the panels to the roof and installed 3 of them on the garage. They put up a large piece of plywood on the side of our house and mounted the 3 inverters, the DC cut off switch and the AC cut off switch on it.

    Thursday, they came earlier than usual, around 9:30, or maybe earlier. Not seeing the Akeena van, I didn't know anyone was here until we saw the ladder was set up again. Scanner talked with John, who was working on the electrical stuff and they agreed the plywood would be painted gray to match the vinyl siding. Later while admiring the 4 panels installed on the front roof, John asked me if I got the feel for how it would look. I commented that I wasn't sure RWC would like the way the corner of one of the panels on the garage appeared to stick out. John called up to Julio on the roof to tell him he had the "fun new job" of moving the rails over by a foot. As you can see, there is space to move the rails. I gave them the check for the equipment as well as the second copies of some forms that I had thought were our copies, but apparently PG&E needs both copies. At the end of the day there were 15 or 16 panels installed. I don't see how they can finish the job in the 5 days they originally estimated, but I figure that's because they were estimating based on a typical project with fewer panels and/or lower percentage of roof space covered.

    Friday they appeared at 9am. With the 8 panels on the front roof, passersby were stopping to ask about them. "How many panels?", "Will they use batteries?", "How much does it cost?", "Wow, that meter is spinning fast, will it really spin backwards?" Yes, it will, every sunny day of the year. The panels on the garage roof were moved so they would not be visible from the front of the house. 8 panels more panels were put up on the South facing roof, which is nice, since the plan showed only 7 at that location, while a String is 8. Nice to have them all lined up in one spot. A conduit was installed through the roof to the switchbox, and a bunch of wiring was done. 20 panels on the roof at the end of the day. John had to leave early at 3pm to get ready for a trade show. Julio left at 4pm.

    Saturday Julio and another installer spent a couple of hours of overtime and put up 3 more panels.

    Monday they came at 1pm, since their weekly Monday morning meeting ran late. They took measurements of the back roof where 15 panels will be fit "like a puzzle", and marked off in red where the rails and mounting hardware would go. Some more wiring went on at the side of the house.

    Tuesday they showed up bright and early around 7:30am and brought the remaining 16 panels with them. They put the brackets in the back roof, and then the rails, and then the clamps. Suzanne, the office manager showed up to see how an installation goes. It was hot out, as was Monday. Barry, the President of Akeena dropped by to see how things were going, but it seems I missed him again. Then around 2pm, Gabe showed up to help with the installation. The newly delivered panels went up on the roof. Seven were installed on the bottom row where the drawing only shows 6. The 7th one is visible from the street. In total, 30 of the 48 panels are installed.

    Wednesday, two started working at 8:30am. Looks like they did some wiring of the panels and then started questioning how to get 2 strings of 8 panels in the south facing direction, considering they had fit 8 panels instead of 7 on the bottom row and they had not reverse tilted the panels on the small north-facing roof. It would be great if they could fit 4 panels on the top row instead of 3. They put in some brackets on the final north roof and left at 3pm saying it was too hot.

    Thursday they arrived before 8am again and put up rails and 4 of the 8 panels on the north roof. They managed to fit the 4 panels on the top row of the south roof, so they removed the no-longer needed rails and panels from the small north-facing roof. They also removed the panel and rails from the small west roof, which I take to mean they feel they can fit all 16 panels on the large west-facing back roof. The shifted the 7 panels on the lowest row of that back roof over to so the last panel would show as little as possible from the street. A lot of ground wiring went up, and the Sunnyboy Control unit was installed.

    Friday they put up 7 panels on the middle row and 2 panels on the top row on the back roof for a total of 16 on that roof. The top panels are just barely visible from the street. The last of the 7 panels on the middle row is more visible from the street. After putting up the 8 panels on the north roof, they trimmed off all the excess rail. With all the panels installed, only wiring should be left.

    Memorial Day, they actually came and worked almost the entire day on wiring. Tuesday they were back again for another hot day that hit 90F by the time they left at 2:45pm. They said they will be back for two more days to redo some of the wiring to make it safer and add some more junction boxes. I must say I'm a bit puzzled, considering all last week even on Friday, they kept saying they thought they'd finish that week. Suddenly there's four extra days. Still, if it's for safety I'd rather have it done right even though it takes longer

    Wednesday they finished up on another 93F day. They called the city to request an inspection on Friday, although since they called before 3pm, they actually could get the inspection on Thursday.

    Apparently the city inspector came unannounced on Thursday, which we didn't find out until John from Akeena showed up on Friday and called the city to find out what was going on. The inspector marked off a couple items on the permit, including some labelling which Akeena had planned to do on Friday while waiting for the inspector to show up. Friday, May 30, John did a brief commissioning test run. Despite it being an overcast morning, I could see our electric meter noticeably slow down. He also showed us where the fuses were in the unlikely case they ever blew.

    Monday, promptly at 9am, the Redwood City inspector arrived. Seeing no changes, he was about to write up the same items as on Thursday. Fortunately John from Akeena arrived a few minutes later with the labels and a 50A breaker while the inspector was still there and got a signoff on the permit. He also sealed the joint on the EMT conduit on the AC side and tightened some joints. There was also talk of a lightning arrestor. Many locales require this, but Redwood City does not. We rarely have lightning storms.

    In all the excitement I forgot to give John the check for completion of the installation and the permit signoff. I can mail it in.

    Official Production

    I called Akeena later in the week to find out when PG&E would come for an inspection. Apparently there was a bit of miscommunication and PG&E hadn't been notified yet, because the right person hadn't received the building permit signoff yet. But that got straightened out and we waited for the call from PG&E. Meanwhile the usually monthly meter reading on the 18th came and went.

    Finally we got the call and the PG&E guy came out on June 20. He asked us to turn off our lights and computers and other power-using appliances. Then he had us turn on our solar system. It takes about 8 minutes for the inverters to come up and start feeding power into the house. We watched as the meter started turning backwards. He then threw the main house breaker off to verify that when there is no power from the grid, the inverters also shut off power from the PV array. This is to ensure that during power outages, the PV array does not feed power into the grid endangering utility workers. It worked just fine.

    Then he had us turn the solar system off. He was surprised at the meter was still spinning forward at a decent speed. We figured it was because off all those devices that always on, even when they are off, such as all the A/V equipment, the UPSs, things with clocks in them, such as the microwave, and the "wall-wart" chargers.

    We both read the old meter before he swapped out our old meter for a new meter which was guaranteed to measure power usage accurately in both directions, running foward and backwards. I take this to mean that while our old meter ran backwards and seemed accurate enough to me, it was not guaranteed to be accurate when running backwards. He had us turn our solar system back on and he put a tag on our new electric meter stating that our house is approved to interconnect our power generating system to the grid. It's official! Just in time for the first day of summer, the longest day of the year.

    The First Month

    The first month, we generated 1376kWH, averaging almost 46kWH per day. Not quite as high as predicted, but quite nice. We reached as high as generating 49kWH on our best day.

    We received an electric power bill for the two days between the last regular meter reading and when the new meter was installed. Curiously, our power consumption has gone down, not even considering the solar. This is likely due to decommissioning two of our oldest power-hungriest computers, and the failure and disconnection of one of our old UPS (uninterruptible power supplies)

    After a full month we got our first bill under the net metering. Instead of the usual small envelope, it was a 10"x13" envelope with the usual small bill enclosed, and seven 8.5"x11" sheets of details relating to the net metering bill. Basically it charged us for the $.45733/day rate we have for just being connected on the E-8 schedule plus the related legislated reductions, surcharges and taxes. This came out to $12.11 for 28 days for us. We actually generated 73 KWH of energy more than we used. This was listed as a "Current Unbilled Credit is: $-8.63". It's a nice thing to see.

    It's ironic, since we used less power than we historically have, we only saved $231 on our bill vs $317 if we had continued to use 80KWH/day. This means that our payback period on the solar panels will be even longer. But I'm not looking a gift horse in the mouth. A smaller electric bill is a smaller electric bill.

    For the month of July, we generated 1325 kWH of energy which is 9% less than the 1459 I predicted. Some of this might be because the trees in our back yard have grown taller and do cast a shadow on our west-facing panels in the late afternoon. Some of it might just be natural variations. It should help if we rinse off the panels which have gotten dusty.

    Since we do not have a time-of-use meter, I don't think we can ask PG&E for a rate comparison of E-8 vs E-7 for us after 12 months. However, Scanner did install a device that will send our house power usage information to his computer. Combined with the data from the Sunnyboy Controller, we should be able to do the rate-schedule comparison ourselves.

    On August 8 we received the $4/watt rebate check. Not bad. That's less than the 2 months we were warned it might take for PG&E to cut the check. It's nice to get a $29,472 check. Now we can pay off the home equity line of credit.

    I noticed the check was smaller than I expected. Somewhere between February, when I first calculated the rebate, and June when Akeena submitted the paperwork for the rebate, the CEC changed their inverter efficiency ratings from 3 significant digits to 2. So the CEC listed efficiency for the Sunnyboy inverter went from 94.4% to 94%. That missing 0.4% has cost us $126 worth of rebate. That just adds a few more weeks to our payback period.

    UPDATE 2005 - I no longer think we actually used less energy after putting in the solar arrays. For one thing, another UPS died in 2005 and I did not see a significant drop in our power usage. I also got a Kill-a-Watt meter in 2005 and found that the percentage difference between VA and RMS Watts is about the same as the percentage difference in our KWH/day before PV vs after PV. A friend who has some knowledge of electric meters told me that a lot of the old dial-type meters measure VA, while the new digital meters measure RMS Watts. Suddenly I get the feeling that we've been over-paying for electricity for years.

    The First Year

    During our first year on the net-metered billing, our solar panels generated 10,536KWH of energy. That's 7.5% less than Akeena originally estimated. This is not surprising since they weren't assuming that any of the panels would be facing north. The idea was that they could be reverse-tilted. However Redwood City didn't like the reverse-tilt idea. The 10,536KWH generated was also 6.5% less than the 11,270KWH I estimated using the nrel PVWATTS calculator. We still haven't tried calculating how much of this is caused by the shade from our neighbor's tree in the late afternoon.

    We saved $2,010 in that first year, compared with what our electric bill would have been if we had no solar panels. This is 23% less than Akeena predicted, but besides the 7.5% lost because of the north-facing panels, the prediction is off because it was based on our using close to 80KWH/day which would put is in the top billing tier with its 12.505 cents/KWH surcharge. If all the energy the PV panels generated had gone to offset top-tier power usage, we would have saved $2,390. We actually averaged 45KWH/day in energy consumption.

    Our electric bills from 5/21/2002 to 5/19/2003 totalled $5,517. Our bills from 6/18/2003 to 6/17/2004 totalled $817, $668 of that was for the net energy used, $149 was for just being connected to the grid. So compared with the previous year we saved $4,700. As previously mentioned $2010 of that was savings from the PV production which means $2690 was from conservation, or using less electricity. Update 2005 - or from the old dial meter overcharging us based on VA instead of Watts.

    If we had been on an E-7 billing schedule, we could have saved roughly another $500 which would have brought our annual electric bill to a total a little over $300. However, I knew we were getting a jacuzzi (it was installed on June 3), which increases our power consumption and I want to make sure that the E-7 schedule will still make sense before I change to it. Once we change to E-7 we can't change back to E-8.

    Now that I've done the analysis, I realize I didn't have to wait. I could have figured it out 6 months ago when I had a full half year's data from the longest day to the shortest day of the year. It is highly unlikely that we'd ever use enough additional power to wipe out the $500/year E-7 vs E-8 savings. If we increased our power usage by 35KWH/day, from our current 45KWH/day to our previoius 80KWH/day, and if that power were spread out evenly throughout the day (as uninterruptible power supplies (UPSs) and computers always on might be) then we'd lose about $168 of that $500 savings. We'd still save $332/yr with the E-7 schedule. To wipe out the $500 savings we'd have to either use 104 more KWH/day than we do now (spread evenly throughout the day) or use 11 more KWH/day during peak hours (that's like having 2670 more watts on continuously during peak hours).

    The Second Year (TOU)

    Towards the end of July 2004, I called up PG&E customer service and asked to switch to the E-7 Time-of-Use schedule. The lady on the phone offered to send me a rate comparison to see if changing billing rate schedules would save us money. I was surprised. I didn't think they'd have the data to be able to send me such a comparison, but I said, "Sure!". While trying to send that out to me, she realized she couldn't, and that I had to call the Business Customer Center, whose number is on the multipage 8.5"x11" bills I get once a month right after the smaller residential bill arrives.

    So I called up the Business Customer Center and they said I needed to fill out a new form and send a check for $277, and no they couldn't send me a rate comparison. The form arrived in the mail the very next day! I sent in the form, which came back the next week with their signatures on it.

    I wondered what was next? Would they call to schedule a day to switch the meter? No, they just showed up one morning, 10:40am, Tuesday, August 3rd. I suppose they would have knocked on the door first before switching the meter, but Scanner happened to be walking out the door just that minute. We shut down as much as we could in the house (other than phantom loads and such), turned off the solar system, and they swapped out the meter. It was nice to see after just 4 hours that we had already sold 14 KWH of power to PG&E. (50000 on the meter is what we'd normally think of as 0.)

    A couple of days later the bill detailing the $277 showed up. $152.96 was for the meter, CIAC tax on that was $52.01, which comes out to 34%. Wow! "CIAC" seems to stand for "Contribution in Aid to Construction" and is some kind of Federal tax. The remaining $72.03 was for installation costs. Since I expect that the E-7 schedule will save us about $500/year more than the E-8, the $277 should pay for itself in about 7 months.

    Another thing we've noticed, is that spraying the solar panels with water is not enough to clean them after months of no rain. We've been using a squeegee to help clean the panels we can reach from a ladder.

    The first E-7 TOU bill showed up. I noticed that the surcharge rates dropped for usage above 130% of baseline. This is one of those gifthorses not to look in the mouth. The good news is slightly lower bills, the bad news is that it makes the financials of the PV look less attractive. Adding the effects of the lower surcharge rate and us going to the E-7 schedule, I think we come out ahead. (The lower surcharge rates has something to do with the politics involved with PG&E getting approval to get out of bankruptcy. Someone demanded that the ratepayers get something out of the deal.)

    We had the meter swapped out half way between two regularly scheduled meter readings. For the last two weeks on the E-8 schedule we had a "current unbilled charge" of $30.97. For the first two weeks on the E-7 schedule we had a "current unbilled credit" of $20.15. We saved $43.50 with the E-7 vs the E-8 in those two weeks alone. It's always nice to see a credit.

    The Third and Fourth Years (Games with Baselines and Surcharges)

    Surcharges started going up again in June 2005 and then October. In January 2006 it rose above the June 2001 surcharge rates and in March they went even higher. Now on the E-7 TOU schedule, in the summer during peak hours, the highest tier rate is 50.5 cents/KWH.

    I have mixed feelings about this. On one hand our electric bill is going up, and on the other hand, that means the solar panels are paying for themselves faster. While this was expected, I had chosen not to depend on it for my payback calculations.

    Hidden among all these changes in surcharges, they added in a change in calculation method that I don't have mixed feelings about. They are definitely adding to my bill and reducing my savings by artificially lowering my baseline quantities and thus increasing my surcharges. Details at When Net Metering is Not Truly Net Metering.

    I sent a formal complaint letter to PG&E with my calculations of just how much I believe they overcharged me and how much they decreased my Total Baseline Quantity. If they say that the CPUC approved the changes, I want it in writing. I haven't found anyone else who says the same thing has happened to them, but I suspect that it's not because I'm the only one. I think it's because the 16 page bills are difficult to verify. I take http://debris.com/journal/1453 as evidence that it's happened to this solar blogger. A $400+ error was caused by one meter misreading followed by a correct reading. There wouldn't be such a huge error if it were truly net metering.

    PG&E sent me a letter acknowledging that they received my complaint. A month later, I still hadn't heard from the CPUC.

    As of May, customers can no longer switch to the E-7 schedule, but those of us already on it, can stay on it. Instead they have reopened the E-8 Seasonal schedule and opened an E-6 Time-of-Use schedule that has peak hours, partial peak hours and off-peak hours. I did some calculations and found that the E-6 schedule would cost us more than the E-7 schedule.

    Two months after I sent in the letter complaining about the changes in billing calculations, I got a phone call from the PG&E executive office. They apologized for the delay. I was pleased to hear that I was not the only customer to complain and that PG&E's tariff department agreed with me that the billing is incorrect. Since they are a regulated utility, they have to file a petition with the CPUC to change the billing calculations and the CPUC has to approve it. Then PG&E will go through the billing records back to October 2005, look at all the customers and credit the accounts accordingly. This should happen by the end of the year.

    Meanwhile, I was promised a letter in writing explaining what happened. The verbal explanation didn't make much sense to me. Yes, I understand that the new calculation method works fine for most people, and is only a problem for those of us who generate power and send that power back into the grid from time to time, but that does not explain why a perfectly good calculation method before October 2005 was changed at all.

    For our July 19 to August 16 bill, I noticed that the baseline calculations have reverted back to the old correct way. This is different from the change that PG&E said they'd do in the letter they sent me.

    Per our conversation today, PG&E is in the process of filling (sic) with the CPUC to "requests authority to revise Special Condition 2 entitled 'Net Energy Metering and Billing' to address baseline allocation where a net energy metering (NEM) customer's net usage for all time-of-use periods totals zero (i.e. when net generation in one or more periouds exactly offsets the net usage in all other periods), then the value of usage and/or generation will be calculated using Tier 1 rates (as set forth in the otherwise applicable rate schedule.)
    Considering that there were positive and negative usage values in my bill for Tiers 2, 3 and 4, this does not appear to be the result of the new filing.

    I wonder instead if this is the result of more games with baselines and surcharges. On August 15, PG&E issued a news release stating

    The utility plans to provide a one time bill credit of 15% for all residential customers... The prolonged regional heat wave in late July caused a spike in electricity consumption as customers were struggling to stay comfortable in the blazing temperatures. Comparing July to June, PG&E residential customers' electricity usage increased by an average of 28% per customer from 537 kwh to 690 kwh as a result of the high temperatures and average bills increased an average of 44%, from about $79 to $114....The proposal enables PG&E to accelerate the credit that customers would normally receive at a later date. PG&E anticipates that the total amount credited to customers in October will be approximately $125 million.

    Our bill for the month covering the heat wave was $71 for the electricity. If they had calculated it the same way that they had for the previous few months, it would have been $101. By charging us correctly that month, they get to send us a smaller refund check in October. If they go back to the incorrect way of billing the next month, it will be very hard to convince me that they aren't playing games with baselines for their financial benefit.

    Well, they didn't go back to the incorrect way of billing, and we got our October refund check, but we have not ever gotten a credit for all the months that they over billed us.

    In February 2007, suddenly the usual 17 page bill became a 2 page bill with extremely few details. At least their dollar amount for the month matched mine. But are they preparing for more funny business with billing and making it so there's no way for us to see what their mischievous calculations are?

    The Fifth, Sixth and Seventh Years

    Despite the bills with no calculation details, there has not been any more funny business these past few years. My calculation for the bill come to within a few cents of their bill. The differences can be attributed to rounding errors.

    In 2009 electricity rates went up again. The highest tier rate on the E-7 schedule went to 61 cents per KWH during summer peak hours. In the sixth year, 2008-2009, we saved on our electric bill 42 cents per KWH generated by the PV array. In the seventh year, we saved 47 cents per KWH generated.

  • First Year Data
  • Second Year Data
  • Third Year Data
  • Fourth Year Data
  • Fifth Year Data
  • Sixth Year Data
  • Seventh Year Data
  • Current Year Data
  • Multi Year Data
  • Links

    Comparison of the systems

  • Sun Power Geothermal Proposal
  • Battle for Business

    Federal and all States Incentives for Renewable Energy

    California Incentives for Renewable Energy

  • $4 per watt California rebate (decreasing in steps, currently $1.55/watt)
  • Eligible PV modules and their CEC PTC rating
  • Eligible inverters and their efficiencies
  • 15% California tax credit (decreased to 7.5% and now expired)
  • California Solar Rights Act Section 714 of the Civil Code


  • Clean Power Estimator for California, New Jersey, Florida, Hawaii, New York, Long Island, North Carolina, Ohio and other states (I highly recommend it)
  • Performance Calculator for PV Systems in the US or maybe just words talking about it and no link to the calculator
  • SMA String Sizing
  • Cooperative Community Energy estimated energy output

    Solar Contractors in the US CalSEIA solar contractors in the San Francisco Bay Area:

  • Sun Power & Geothermal Energy, now SPG Solar
  • Akeena Solar rebranded Westinghouse Solar
  • SolarCraft Services
  • EcoEnergies/PVI Construction (now REC Solar)
  • Light Energy Systems (acquired by Acro Energy)
  • Sky Power Systems
  • PowerLight Solar Electric Systems acquired by SunPower Corp
  • Occidental Power
  • MC Solar Engineering
  • Next Energy Corp
  • Sun Light & Power


  • BP Solar
  • Arizona Solar Center

    Efficiency records
  • Efficient solar panels in production 20.4% efficient SunPower 333W module 2009. 19.1% efficient SunPower SPR-315 module 2007. 18.3% efficient (2005) (downgraded to 17.7% efficient in 2006?) Sunpower SPR-220 module. 16.9% efficient Sunpower SPR-210 module October 2004. 15.7% efficient Kyocera module Aug 2004.
  • Efficient thin film solar modules in production 10.7% efficient Global Solar Feb 2004. (If their website is unfriendly, go to their homepage and try again.)
  • Efficient thin film solar modules in pilot production 13.5% efficient Shell Solar dead link Nov 2005. Copper-Indium-Diselenide (CIS) 9.3% efficient in production
  • Efficient solar cells in production 39.2% efficient terrestrial solar cell from Spectrolab 2010, 28% efficient from Spectrolab
  • Efficient solar cell (1 sun, not high concentration) 42.8% efficient 2007.
  • Efficient solar cells under test 32% efficient from Spectrolab test at Arizona Public Service Nov 2004.
  • NREL tested record-setting efficent monolithic solar cell 32.3% efficient from Spectrolab Nov 1999.
  • NREL tested record-setting efficient concentrator solar cell 41.6% efficient Aug 2009. 41.1% efficient Jan 2009. 40.7% efficient 2006. 37.3% efficient 2004. 36.9% efficient 35% efficient lattice-matched, triple-junction, GaInP/GaInAs/GE cells at 66-suns concentration 2003.
  • research samples efficient thin film solar cells >20% efficient Australian National University 2006.
  • NREL tested record-setting efficient thin film solar cell 19.9% efficient CIGS 2008. 19.2% efficient copper indium gallium diselenide 2005. 18.8% efficient 1999.
  • most efficient concentrated photovoltaic (CPV) solar power receiver 30% efficient Solar Systems CPV 2006
  • NREL Best Research-Cell Efficiencies graph

    Non-traditional technologies

  • Concentrator PV flat panel

    Custom Production

  • BIPV windows by Energy Photovoltaics Inc
  • Future technologies

  • Active Building Envelope Thin film technology that adheres both solar cells and heat pumps onto surfaces. RPI.
  • Nanorod solar reseach by BP & Caltech
  • Carrier-multiplication-enhanced solar cells
  • Holographic Solar
  • Concentrator PV flat panel
  • Glass windows that produce electricity article
  • nanotechnology - silicon quantum dots.
  • 30% efficient flexible plastic solar (infrared) cell University of Toronto
  • low cost organic solar cells
  • high efficiency solar window shades Rensselaer Polytechnic Institute update?
  • low cost high efficiency solar technology

    Other interesting pages

  • How Solar Cells Work
  • History of photovoltaics
  • PV education
  • PV system design and installation
  • North American Board of Certified Energy Practitioners resources
  • Rocky Mountain Institute
  • US DOE Energy Efficiency and Renewable Energy - Solar Energy
  • Solar Timeline
  • Union of Concerned Scientists on renewable energy
  • Renewable energy stocks
  • Alternative Energy News

    last changed 2009.12.21

    Carolyn Luce