While one half of the roof is dedicated to generating hot water, the other half is our electric power plant.
A roof has traditionally been designed to protect the interior of a house by keeping the weather, birds, leaves, etc. out. It is usually the largest surface area as well. Why not put it to work?
In our case, the roof has been "solarized." When you start from scratch, the architect can meld the various functions into the overall design. There are no shingles--actually, it's shingled in glass which covers the solar thermal panels and is part of the photovoltaic panels.
This is a view of work on the south-facing roof taken in the late fall. It took about three weeks to ever-so-carefully get it in place. Only the leaves fell.
Sixteen 4'x6' panels provide 4200 watts of DC power at 48 volts to an inverter in the basement. They are essentially designed for commercial use and were built by ASE of Billerica, MA (formerly Mobil Solar). The inverter's output is 240 volts ac. We anticipate generating in excess of 4 megWhrs on an annual basis.
During our first day hooked up to the sun, the inverter's output reached 3.9 kW (we came online at 12:30 p.m., March 3, 1995). Several days later, the digital meter displayed 4.01 kW - only a 5% reduction from the rated power of the panels. The diagram to the right is a modification of a graphic from Michael Potts.
Central Maine Power installed two meters, one to measure consumption of grid power and the other to register our excess production. The consumption meter reduces its spin-rate noticeably or stops completely when our inverter's output is added to the mix. Maine is an "annualized net metering" state which means any surplus generated on a monthly basis goes into a 'bank' that we can use during the lean winter months. In 2001, the 'bank' had more than 500 kWhr available for late fall/winter use. We essentially pay nothing for electricity...just a monthly hookup charge of approximately $8.00.
Our first Trace inverter was installed near the end of September, 1996 after the previous inverter (by another manufacturer) failed four times in a year and a half. In my book it's "four strikes and you're out!" The Trace was a pure sine wave inverter (SW5548UPV) capable of handling the 4200 watt input from the PV array. It weighed around 150 pounds and that made lifting and attachment to the plywood wall a bear of an event. In several maintenance trips to Trace, it suffered the indignity of a bent frame and balky internal fans. It has served us well (still works) but it was time to move on to a newer model that promised greater flexibility.
We upgraded our conversion of DC to AC on September 8th, 2001, by installing two Trace Sun Tie Inverters. These inverters were a disappointment - failing to live up to expectations.
On September 30, 2002 (one year later), the Sun Ties were upgraded to the XR versions and to date they are a considerable improvement.
On our building plans we located a site for a wind generator - in case we want to expand our generating capacity. During the winter, there is quite a brisk nor'westerly that would more than compensate for that season's reduced solar gain. However, since I am in surplus on an annual basis, there seems no point in adding a wind turbine.