| Once you've decided to add a system of solar | | | | the sun, the number of hours of sunlight your roof |
| photovoltaic power panels to your house or business | | | | receives is not the same as the hours of direct |
| or determined that it's time to expand an existing | | | | sunlight, since much of that light will probably not be |
| system, the next question is: how much wattage do I | | | | coming in at right angles to your roof throughout the |
| need? | | | | year. |
| To find this out, you will need to know how much | | | | Another way to estimate the number of hours of |
| power you consume (or how much you plan to | | | | direct sunlight you receive is to assume a value or to |
| consume, for new buildings or for additions that will | | | | use a rule-of-thumb. Often, 5 hours direct sunlight on |
| require lots of power - like a woodworking shop or a | | | | average throughout the year is assumed. Although this |
| hot tub) and how much sunlight you receive throughout | | | | is a fairly common assumption, you might end up |
| the year. | | | | disappointed if your solar system generates less |
| Assuming that you've been at your location for at | | | | power than you anticipated because you get fewer |
| least a year (so that you have a good feel for your | | | | average hours of sunlight than this. |
| electric needs for each season), take each electricity | | | | A third method is to look up an average regional value. |
| bill and write down the number of kilowatt-hours (kw-h) | | | | The Florida Solar Energy Center presented a study in |
| that were billed and the number of days in the billing | | | | August 2004 at the ACEEE 2004 Summer Study on |
| cycle. Then, divide the number of kw-h by the number | | | | Energy Efficiency in Buildings titled "Geographic |
| of billing days to get the average kw-h per day. To | | | | Variation in Potential of Rooftop Residential |
| make things simpler, some people just assume 30 | | | | Photovoltaic Electric Power Production in the United |
| days for each billing cycle (and not all power | | | | States." This study simulated the power generation of |
| companies report the billing days). That's how many | | | | a 2 kilowatt (kw) photovoltaic rooftop system for 236 |
| kw-h your system will need to generate each day to | | | | locations throughout the US. Depending on the location, |
| meet your power needs. Once you've done this for | | | | they found that this 2 kw rooftop system would |
| the entire year's worth of bills, you'll know both your | | | | generate between 5 and 8.5 kw-h per day on |
| average daily consumption and your peak | | | | average throughout the year. |
| consumption. You may choose to size your system | | | | Using the rule-of-thumb, a 2-kw system receiving direct |
| based on either value. If you use the average value | | | | sunlight for 5 hours per day would expect to generate |
| (and assuming your power company allows you to sell | | | | 10 kw-h per day on average throughout the year. |
| power back to them, you should definitely check on | | | | From the Florida study, the same system would be |
| this first!), then some days you'll make extra power | | | | expected to generate only 5 - 8.5 kw-h per day |
| and sell it back to the power company and some | | | | throughout the year. |
| days you'll be buying power, but over the course of a | | | | How you proceed depends on your goals. If your |
| year, your net electric bill should be close to $0. If you | | | | goals are to offset your electricity cost, then any of |
| use the peak number, then you should be selling power | | | | the approaches will provide an estimate of the number |
| back to the power company almost every day, but | | | | of hours of direct sunlight that you receive. However, if |
| your system will cost more than if you sized the | | | | your goal is to become energy independent, then using |
| system based on the average power consumption. | | | | the results of the Florida simulation would provide a |
| Now, to get the wattage of solar panels that you | | | | more conservative value that would be more likely to |
| need, you'll need to know how many hours of direct | | | | provide a system sized for your needs. Take the |
| sunlight that your system will receive. If you've been at | | | | results of the simulation and a few of your own |
| your location for several years already, you might | | | | observations (whether you have a lot of tall evergreen |
| have a good idea how many hours of direct sunlight | | | | trees on the south side of your house that block a |
| your building receives, either seasonally or on average | | | | majority of the winter sun, etc.) and you should be able |
| throughout the year. One way to determine how many | | | | to estimate the number of hours of direct sunlight that |
| hours of sunlight you receive is to take observations | | | | you receive. |
| several times a day as to whether the portion of your | | | | For example, if you've decided to use the average |
| roof where the solar panels are to be installed is sunlit. | | | | power consumption value and you use 600 kw-h per |
| If you do this every day for a year, you'll have a good | | | | month and your billing company doesn't report the |
| idea of how many hours of sunlight your solar panel | | | | number of days per billing cycle (assume 30 days per |
| will receive on a seasonal basis. There are a few | | | | month), then you consume 20 kw-h per day of |
| problems with this approach: it's very time consuming | | | | electricity (600 kw-h divided by 30 days per month). |
| as it requires you to take measurements throughout | | | | Using the Florida simulation data and assuming you live |
| the day nearly every day; there will be some days | | | | in Western Kentucky, you receive approximately 3.5 |
| when you just can't take the measurements (you're | | | | hours of direct sunlight per day (from the graphic, |
| out of town, working and cannot get to the house to | | | | estimate a value of 7.0 kw-h divided by 2 kw - the |
| observe the roof, clouds are obscuring the sun, etc.); | | | | study is based on a 2 kw system). Then, your system |
| and, unless you have a solar panel system that tracks | | | | would need to be sized for approximately 5,700 watts. |