Regarding this article, I think I’ve found a simple reason for this failure, and the reason will shock and surprise you.
In the article, a news report by TV station WFTS is cited:
WFTS News in Tampa obtained copies of the courthouse’s electricity bills and confirmed the savings are no more than about $2,000 per month. WFTS also confirmed the panels are reducing electricity bills by only 15 to 18 percent, instead of the promised 40 percent.
You can watch the news report here.
They say a picture is worth a thousand words, I’ve got one of those. First, look closely at the picture in the news story above…now look at this picture and note the arrow.
I think the government dufuses and the solar company missed the shading from the nearby tall building (the Hillsborough County Center Building), seen in the photo above which you can inspect yourself at Bing Maps here: http://binged.it/1tqPjnf
It looks like a little over half of the panels would be in the shade during the afternoon based on my comparison to the article photo, a larger version which exists here:
That afternoon shade will kill solar panel efficiency big time. The problem will be greater in the winter, at low sun angles, further reducing the efficiency of solar panels which appear to be placed flat on the roof, rather than tilted for maximum efficiency. In fact if you watch the WFTS i-team video, you’ll see that the panels are in fact laid directly on the roof surface. Here is a screen cap showing workers placing flexible panels flat on the roof:
The effect of array tilt angle on solar PV energy output may be up to 20% compared to that of flat installations, depending on latitude. Typical rooftop arrays with tilt look like this:
You can read more about tilt angles and panel efficiency here.
But the tilt issue is small compared to the shading issue by the tall building to the Southwest. using Google Earth’s timeline feature, I found an image from Dec 26, 2012 that shows the peak shading near the winter solstice, it also shows the solar panels in place. Clearly, the shadow will cover a significant portion of the solar panels on the roof for a period of time.
The next thing that needs to be done is a public records request to get the hourly solar power system output data over the past couple of years, it should be an easy matter then to note when the output drops significantly, and to line that up with sun angles and times.
On a daily basis, you can watch the real-time output page here:
Below the big dial gauge, if you choose “select gauge” and the “Interval 2-hour Avgs.” you’ll get a bar graph plot of solar output by hour, though there is no facility for getting anything but today’s data. From my observations, it seems that at this time of year, we get a bigger drop-off in production in the late afternoon than we should, which should be almost symmetrical with the buildup in the morning towards peak insolation. The afternoon production seems to have a larger gap.
It will be interesting to see what the public records request for the hourly solar power output data brings.
Full disclosure: I have a solar power system on my own home.