We've Gone Solar!
Installing our 2528 Watt Tracking Photovoltaic Array
By Thomas J. Elpel

      We have been asked more times than I can count if our owner-built, energy-efficient, stone and log home is off the grid and solar powered. Well no, we've always been on the grid, using electricity from coal-fired power plants and local hydroelectric projects. But we made finally made the leap into solar power, and as of January 1st, 2003 we have been producing our own power. We installed a 2528-watt photovoltaic (pv) array on two Zomeworks Passive Trackers to produce essentially all of the electricity we use.

      In a nutshell, the Zomeworks Passive Trackers are built with two large Freon-filled tubes welded from heavy square stock. There is one tube on each side of the tracker, with a small, flexible tube connecting them. Reflectors on the trackers help to focus solar energy on one of the square tubes while shading the other one. The Freon warms and expands, causing it to flow through the flexible tube to the other side of the tracker. That makes the other side of the tracker heavier, causing the panel to move and align itself with the sunshine. This way the panels track the sun all day long without using any electricity and without using any motorized systems that might break down.

      Our pv system is grid-tied, consisting of sixteen 158 watt Kyocera solar panels, such that in the daytime we run the meter backwards, powering other people's homes with our solar electricity, and at night we draw off the grid, using coal-fired power. On average we generate about as much electricity as we use. In this article I'll lay out all the details about the costs and benefits of going solar.

      The first thing that everyone wants to know about solar power is how economical it is, because they are tired of paying high utility bills every month. They want to install solar panels, generate their own electricity and save money. Unfortunately, it doesn't work that way, at least not yet. Although the cost of solar power continues to fall almost as fast as the price of computer hardware, the reality is that solar remains significantly more expensive than grid-supplied power under most circumstances.

Going for Efficiency
      The single biggest factor in making solar power feasible is still tackling the problem from the demand side, instead of the supply side. Basically, it is impractical to supply the full power needs of a conventional American home with photovoltaics. However, by finding creative ways to reduce demand as much as possible--especially through energy efficiency and alternate fuel sources--you can reduce your electrical needs down to a point where the cost of a photovoltaic system is at least bearable. The idea is to look for ways to eliminate the need for electricity without sacrificing comfort, as detailed in my book Living Homes: Stone Masonry, Log, and Strawbale Construction. Then you may be able to afford solar power for whatever is left that will not run without electricity.

      Turning a house into an energy efficient home is a lot easier when building from scratch, than trying to retrofit an energy-wasting beast later on, but either way is definitely worth doing. We did a little of both in our house, building it as energy efficient as we could with our limited funds, living on a total income of about $12,000 a year (early 1990's). Read my article Building a House on Limited Means for more details.

      Household heating and cooling is usually the single biggest energy expense in a home, so we made sure our place was well-insulated, and we built it facing mostly south with lots of windows for passive solar heating. The winter sun shines in through the glass and helps warm the house, and that is hard to beat. We also built an efficient fireplace for backup heating with wood, so we do not use any electricity for heating. Summer cooling isn't an issue here in the mountains of Montana where the temperature rarely exceeds 90ºF.

      Heating water is typically the next biggest energy expense. We built a solar water heater to provide free hot water whenever the sun is shining. This point may require some clarification, since many people imagine using solar electricity to heat water.

      A solar water heater is much like passive solar heating in a house. Sunshine comes through a window and heats your house or heats water in a tank or in pipes, and it is really simple and cheap. Using expensive solar electric panels to heat water would be really foolish. To use an analogy, you could say it would be like passing the basketball all the way down the court and back when you were already standing under the basket ready to shoot. Just take the straight shot and allow the warmth of the sunshine work directly without trying to convert it to electricity and back to heat again.

      The rest of our hot water comes from our wood cookstove. My grandmother cooked on a wood cookstove and I grew up around it. It isn't the most energy efficient way to cook a meal, since it really burns through the firewood, but we've always appreciated the pure quality of cooking on a wood stove. It has water pipes through the firebox to heat water, so we always have scalding hot water when the fire is going. The solar water heater provides most of our hot water in summer, while the wood stove provides the majority of the hot water in winter. We simply take showers when we happen to have hot water. It may be a bit arcane in this day and age, but we like it that way. With this arrangement we do not need electricity for either water heating or cooking on the stove, although we make extensive use of our electric microwave, crock pot, toaster, and hot plate.

      The next biggest energy use in a typical home is usually for refrigeration. Many people with solar electric homes have installed propane powered refrigerators, as well as propane for their water heating and household heating, to avoid the need to generate electricity for those tasks. However, as the cost of solar panels comes down, it is becoming more and more reasonable to purchase the extra solar capacity to run an electric refrigerator.

      We started out with a small, antique refrigerator we rebuilt ourselves. It added about $5 a month to our power bill, but didn't do much to keep the food cold. When we were feeling richer we bought a new, modern refrigerator with a lot more space that also uses about $5 worth of electricity a month, but does a really nice job of keeping the food cold.

      At one time there was a big difference in energy use between high-efficiency refrigerators from specialty markets compared to the conventional refrigerators sold at appliance stores. But thanks to energy guidelines from the government, mainstream refrigerators have improved to nearly match the most efficient models on the market. Ours is a no-frills model without an ice-maker or anything like that, so it is reasonably efficient. We still have a large, older chest freezer, but we moved it out of the house into our unheated workshop so it won't have to work as hard most of the year to keep the food cold. In this case, increasing energy efficiency didn't cost a dime!

      For lighting we use compact fluorescent lightbulbs through most of our house, which use 25-30% as much electricity as an incandescent bulb to produce the same amount of light. We look forward to switching over to the super-efficient LED lights whenever they become available for standard light fixtures.

      In the laundry room we replaced our old, inefficient top-loading washing machine with a super-efficient front-loading model. The efficiency of a washing machine is measured primarily on two factors: 1) how much hot water is needed to wash a load, and 2) how much water is left in the clothes afterwards to be removed in a dryer. Neither factor makes much difference in our electric use, since we heat our water with solar and wood, and we prefer to hang the laundry outside to dry. However, conserving hot water with the front-loading washing machine improves our comfort level by leaving more hot water in the tank for other uses. And the high-speed spin cycle really wrings the moisture out of the clothes so that they don't take nearly as long to dry, or don't consume as much electricity in the dryer when we are forced to use it in wet weather.

      We also have a gravity-fed spring, so we do not need electricity for a well pump. But there are two adults and four children in our home, with lots of lights, computers, radios, and the VCR and television usually turned on-- even when not in use. At present our average electric consumption is about 400 Kwh/month, which is much less than a conventional household, but quite a bit more than most alternative, solar-powered homes. That translates to an electric bill averaging $30 a month for grid-supplied power.

      Regardless of the cost of power, we try to do at least one energy-related project in the house every year to conserve either electricity or wood, or at least to improve our quality of life without consuming more and more power. The newer, better washing machine and refrigerator were two such projects. Insulating or weatherproofing some part of the house a little better is a common project too. Installing our own solar panels to generate electricity was by far the biggest, most expensive energy project we've done yet.

Making the Leap
      There were several factors that drove our decision to make the leap to solar power, including: 1) peer pressure from people asking if we generated our own electricity in our "alternative" home, 2) personal interest to learn about solar power in a hands-on way, 3) the desire to wean ourselves off of fossil fuels to do our part to curb global climate change, 4) we qualified for a grant to pay back part of the cost of our system, and 5) our business grew to the point that we actually had enough money sitting in the bank to pay for our share of the cost. In a nutshell, we finally made it into the middle class income level, but we had nothing else to spend our money on, since we had no mortgage on our home, we drive older cars, and we had few other monthly bills.

      Funding for the grant came from the Universal Systems Benefits Charge (USBC) on our Northwestern Energy power bill (formerly Montana Power Company). Applicants had to be electric customers of the utility. (See www.montanagreenpower.com for more information.) The program was originally administered directly by the company, and we applied for a grant that would have covered the entire cost of our photovoltaic system. We would have "paid" for our share of the cost through labor, by teaching school kids about solar electricity. We might have been awarded the grant, but the residential program was canned just about the time our application went in.

      Instead, the utility developed a partnership with the National Center for Appropriate Technology (NCAT) in Butte, Montana to administer a residential cost-share grant program for alternative energy systems. The new program gave a rebate of $4.50 per watt of installed capacity. The homeowner had to pay the rest. We didn't have enough experience to design our own photovoltaic system, nor the time to learn to do it ourselves, so we asked Oasis Montana, Inc. to put together a package and order it for us. We described approximately what we wanted--two Zomeworks Passive Trackers with as many solar panels as we could fit on them, and a grid intertie system. The helpful staff at Oasis Montana customized a package to maximize the solar capacity on the trackers, with an appropriate-sized inverter for the project. We did our own installation. Our expenses were as follows, not including 30-40 hours of my own labor:

2528-Watt Grid-Tied Photovoltaic Package
Designed by Oasis Montana
(All prices are from 2002.)

Choosing and Installing the Zomeworks Trackers
      A significant part of the cost ($4066.80) went towards the Zomeworks Trackers, and the associated high wind brackets, steel pipe supports, excavation, sonotubes and concrete. The tracking systems increase the performance of the solar panels by directing them towards the sun all day long. However, the added cost of the tracking systems does not justify the added efficiency of the system. In most cases it would be more cost effective to put that money towards additional solar panels and mount everything in a fixed position. This point is especially true in a grant program like this one where the rebate is calculated per watt of installed capacity. At retail cost we paid about $4.66 per watt of capacity and got a refund of $4.50 per watt, so the solar panel part of the project was practically free.

      The reason we chose the Zomeworks Trackers for our solar array was because we did not have a good place to mount the solar panels on the house. This is a good example of the concept of integrated design. We could have saved the expense of the tracking systems if we had built our house with a section of roof at the ideal orientation and slope for maximum solar exposure. Given that we didn't build our house that way, we had to choose an alternate mounting system. We could have mounted the solar panels in a fixed position at an angle to the roof, but that would have looked really bad. Optionally, we could have done a fixed installation on the ground, but that also would have looked tacky. Instead we chose to mount the panels on the expensive Zomeworks Trackers, which are pretty cool, but also dominate the view of our place. From any place in town where our house is visible, the eyes are drawn directly to the solar array.

      We had one other chance to build the solar panels seamlessly into our place. In 2001 we built a small workshop and planned ahead to mount solar panels on it at some future time. We designed it with a barn-style roof, where part of the roof would be at the ideal slope facing due south. The solar panels would have fit the angle and become part of the structure. But midway through the construction we questioned whether we wanted such a tall roof on our little workshop--plus we just got lazy--so we changed the roof to a shallow gable and decided to deal with the solar panels some other way. That decision cost us more than $4,000 by making it necessary to buy the tracking systems. That expense increased our cost to the point that we don't imagine the system ever realistically paying for itself in reduced utility bills. Nevertheless, we just couldn't stand the idea of having the solar panels mounted at an odd angle to the roof, so the tracking systems seemed like the most aesthetic choice. At least we were foresighted enough to include an extra circuit and wiring in the breaker box for the eventual solar array.

      Mostly we have been pleased with the Zomeworks Trackers, although there have been a few surprises. The first surprise was that the support poles were not included in the kits. It makes sense to do it that way to avoid shipping the heavy steel poles, which can be bought anywhere, but it was an extra expense that was not initially expected.

      A bigger surprise was that there was only one pair of shock absorbers included in the kit, so we had to pay extra for an additional pair of shocks to help stabilize the panels in the wind. I do not usually think of our place as being in a "high wind area" because we are largely sheltered by the nearby mountains, so it is often calm here, even when the wind is blowing hard just a few miles down the road. On the other hand, a "high wind area" is relative, and in the greater scheme of things, all of Montana can be considered a "high wind area". When the wind does blow here, it will keep a person awake at night wondering if the roof is going to come right off the house!

      The fall of 2002 was the least windy fall I can remember in many years, which was really nice while installing the track racks and solar panels. However, the wind started blowing hard just a few days after we finished the installation--before we ordered the extra sets of shock absorbers. The track racks rocked back and forth and slammed against the rubber bumpers. I lay awake at night wondering if the whole thing was going to fly apart, littering our place with thousands of dollars of useless solar junk. At one point one of the shocks even unscrewed itself somehow, so that there was only one shock left on the rack. The problem wasn't readily noticeable, except that the rack was slamming against the bumper really hard. It mildly tweaked the steel framing before I noticed it and screwed the shock back in place. Installing the extra pair of shocks on each rack cured the wind problem, and it hasn't kept me awake since.

      I thought the installation directions could have been more detailed, and I think the project would have been difficult for someone without much building experience. There were also a couple of really funky aspects to the design.

      Near the very beginning of the installation process the instructions called for three heavy steel washers to adjust the spacing of the rack. In retrospect, we really needed an additional five washers there for proper spacing, but there was no way to know that in advance, and no way to add more washers afterwards, without taking the entire rack apart. The spacing problem arose when installing the counterweights to help slow the track racks down when they are swinging in the wind. There is a rubber bumper that keeps the track rack from moving too far to either side, and the counterweight is supposed to swing right past the bumper, but it didn't. The directions called for bending the arm of the counterweight if it hit the bumper. Mangling the arm this way seemed like a poor solution to the problem, and it wasn't nearly enough to let the arm swing by the bumper anyway. Since the rubber bumper is just a heavy steel bolt covered with rubber radiator hose, the simplest solution was to cut the bumper bolts a little bit shorter. That cured the problem.

      Since working the bugs out of the track racks, we have really grown to like them. I think of them as "sun mills" as they turn in the sun and generate electricity. On windy days the panels rock gently back and forth, but still track the sun. However, the track racks are definitely less efficient on cloudy days.

      The panels always end the day facing west, and on cloudy days there is not enough sunshine to "wake" the panels and return them to an easterly orientation. The result is that the panels remain facing west through part or all of the day, tilted away from the available solar energy. A motorized track rack would be able to stay on target in these conditions to extract any solar energy that is available. Even a fixed installation, with the panels facing due south, would perform better on these cloudy days. Fortunately, we don't have too many days without at least some sunshine in this part of the world.

Panels, Inverter, and Wiring
      Installing the Zomeworks Passive Trackers included bolting the solar panels in place. It was a bit tricky working with the heavy panels up in the air, but the job was not too bad, since we could at least stand in the back of the truck to do it. It would have been much more difficult to lift the panels into place from a ladder.

      Installing the inverter, combiner boxes, main disconnect and wiring was considerably easier, although additional directions could have helped with the process here too. Basically, the wires from the eight solar panels on the first track rack are fed into a fiberglass box mounted on the pole, then through conduit underground to a box on the pole of the other track rack. All the wires from both track racks are combined in this box and sent out in a positive and negative wire through underground conduit to the inverter, which is mounted in our workshop.

      The 2500-Watt Sunny Boy inverter converts the Direct Current (DC) to Alternating Current (AC) to match the household current. An LCD panel on the inverter shows how many watts are being produced at any given moment, with cumulative totals in kilowatt hours for each day and for the life of the inverter.

      The AC is routed back outside the workshop to a main disconnect that is provided for the power company. That way the utility can disconnect our solar panels to work on their power lines, even if we are not home. From the main disconnect the wiring is routed back into the shop where it feeds 220-volt power backwards into the breaker box and from there to the house or back onto the utility grid.

      We signed a "net-metering" agreement with the power company so that we are credited for the power we produce. If we produce more power in a month than we use, than the surplus remains on the account for any future month when we draw more power than we produce. That way we can use surplus power produced in summer to make up for those short winter days when we are not generating as much.

      There are no batteries in our system, and unfortunately it will not operate unless the grid is working. However, I think that battery technology is still in the dark ages, and I am glad that we don't have to deal with them. At some point we may want to research the feasibility of modifying our system so that we can use it even when the grid is down, but we have never been overly concerned about that issue.

Real World Performance
      With 2528 watts of solar panel capacity we can generate 2528 watts of power under the most favorable conditions. The rest of the time the system generates less power than that. Several factors affect the performance, including atmospheric conditions, temperature, and snow cover--or the lack of it. A system like ours placed in sunny southern California might never reach peak output, since solar gain would be reduced by the thickness of the atmosphere, plus the amount of water and pollution in the air. The panels also work better under cooler temperatures, so a system in a southern location may never exceed 75% of capacity.

      Here in Montana we lose efficiency due to our short winter days, but make up for it with the cooler temperatures, thinner air and crystal clear atmosphere, plus we get the benefit of additional watts reflecting off the snow cover. On one bright, but cloudy day there wasn't enough sunshine to wake up the passive trackers to turn east, but from their westward position they were producing more than 100 watts of power solely from sunlight bouncing off the clouds and snow cover. Otherwise, on a typical sunny winter day with snow cover we produce 2200 to 2400 watts with frequent short spikes up to peak capacity. Once while I was looking at the LCD screen on the inverter, the output even jumped a few watts above peak capacity for about half a second. As we move towards summer we have the benefit of longer days, but slightly lower output, usually between 2000 and 2200 watts on clear days, probably due to warmer temperatures and additional haze.

      Based on our historic power use, we need to produce an average of 13 kilowatt hours per day to cover our consumption. That is about the amount of power we generate on fully sunny winter days. On cloudy days we produce less. With the longer days of spring we can often produce 21 kilowatt hours per day. The question is, can we produce enough of a surplus on sunny days throughout the year to make up for our power consumption on all the cloudy days?

Flipping the Switch

      It was quite the thrill the first time we flipped the switch to watch the wheel on the electric meter start spinning backwards. We were still waiting for some paperwork to go through before we could get our electrical inspection. Then we could sign the Net Metering Agreement with the utility and they would replace our meter with a special Net Meter to credit us for the power we produced. In the meantime--even if we were not being credited for the power we fed into the grid--at least we would be consuming less and thereby lowering our power bill. So I was shocked to find that our next utility bill was even higher than before!

      As it turns out, the utilities had some trouble with customers in the past taking their meter out and turning it upside-down to make it run backwards. So modern meters are made to run forwards regardless of which way the power comes into the meter. The dial was turning backwards, but we were being charged for the power we produced. Fortunately, the meter had an internal counter that registered all the power we put onto the grid. When the Net Meter was installed the utility gave us a credit for all the power we fed into the old meter.

      After one full year of use, it appears that we are producing almost all of our own electricity. We still pay a $4.60 service charge every month to be connected to the grid, but we typically produce enough electricity to zero out on the energy part of the bill. We produce a surplus of electricity through the summer months, then slowly consume that surplus throughout the short days of winter and the cloudy days of spring. In spring we pay about $5 a month for electrical consumption above what we produce. Our electrical use jumps considerably if we go on vacation in winter and leave a small electric heater running in the house while we are gone.

      We could probably zero out on our power bill over the entire year, except that we added another appliance that draws quite a bit of electricity. We remodeled our kitchen and added a high-tech Fisher & Paykel Dish Drawer for washing dishes. It is a nicely efficient dishwasher, but like most dishwashers it uses an electric current to boost the water temperature. Previously we only washed the dishes when we had hot water available from the solar panel or from the wood stove. Now we turn it on whenever we want, so we are sometimes boosting the temperature of the water from frigid to scalding, and that uses a lot of power.

      Moving our chest freezer out of the house into the unheated workshop where it draws less power may be enough to offset the additional power used by the dishwasher. I'm not sure, but time will tell, and we will continue to invest in energy conservation projects at least until we are producing as much electricity as we are using. It is not really about money. It is more a matter of doing our part to reduce global warming, plus it is just a good sport to seek new ways to trim the power bill every year!

The Big Picture
      If you do the number crunching, our solar electric installation doesn't seem to make much economic sense. Given the total cost of the project, it is unlikely that the system would ever recover it's costs. Even with more than half the cost subsidized by the USBC Rebate and the Renewable Energy Tax Credit, our share of the cost could easily take 20-30 years to recover. It may seem like a poor investment on our part, as well as a waste of other people's money to subsidize this technology that doesn't pay for itself. But I see it differently.

      It is generally agreed that all the necessary technological hurdles have been adequately addressed to make PV systems affordable. The only remaining obstacle is the scale of production. There is an effort in California to guarantee a market for a large number of PV systems (at a lower price), provided that the manufacturers are located in California. The thinking is that if you create a large enough market, then PV manufacturers will be able to scale up production enough to bring the unit cost down to greatly expand the market.

      Government meddling in the markets is often viewed suspiciously, as subsidizing products that cannot make it on their own. But there is precedent here. The U.S. Department of Defense was about the only purchaser of the first silicon chips in 1961. The chips from Texas Instruments cost $100 each and didn't do much. They replaced only a couple dollars worth of conventional electronics. Many people considered it a foolish waste of money. But the government orders created enough demand to bring the market price down so that private companies also started buying the chips. The cost per chip fell to $2.33 by 1968 and sales to the private sector quickly dwarfed those to the government. Today we have low-cost silicon chips for personal computers that have far greater processing capacity than the sum of all chips put together at the time of the first trip to the moon. If the government didn't take part, then the computer revolution may not have happened yet, and you wouldn't be seeing this web page now.

      Similarly, you could say that the only reason we could afford our PV system is because someone before us was foolish enough to pay a whole lot more for their system than we did for ours. In this respect, you could say that PV systems are generational. Ours is the result, the offspring, of an earlier, more expensive system, and ours will be the grand-daddy of other PV systems to come. In other words, the payback period might not be great for us on our system, but by installing 2500 watts today, we've done our part to bring costs down so that other people will inevitably install 250,000 watts of new systems at a lower cost a few years down the road. Suddenly it looks like we got a really good deal for our money!

      Markets do not readily recognize the value of benefits to "the commons", but I do. Our PV system helped to make many other PV systems more affordable at a future date, which cumulatively goes a long ways towards preventing the planet from boiling over from fossil fuel emissions. That is very much in my own self-interest, even though I live in Montana where even the warmest years are still on the cold side!

      Looking for life-changing resources? Check out these books by Thomas J. Elpel:

Green Prosperity

Roadmap to Reality

Living Homes

Participating in Nature

Foraging the Mountain West

Botany in a Day

Shanleya's Quest

Return to Thomas J. Elpel's
Web World Portal

Thomas J. Elpel's Web World Pages
About Tom | Green University®, LLC
HOPS Press, LLC | Dirt Cheap Builder Books
Primitive Living Skills | Outdoor Wilderness Living School, LLC
Wildflowers & Weeds | Jefferson River Canoe Trail
Roadmap To Reality | Search this Site!
© 1997 - 2023 Thomas J. Elpel