One Mega-Watt-Hour (1 MWh) electricity from the sun

1000 kWh == 1 MWh

Our solar panels were installed on the 5th of April this year. This afternoon, after just 73 days of operation, the gauge on the inverter reached the point of 1000 kWh, or 1 MWh of production. That's a quarter of our consumption of electricity last year, and as we've only actually used 662 kWh in the last 73 days, our production of electricity has been more than 50% greater than our use over this first 73 day period.

In 73 days, the inverter has been active for 1119 hours. That's an average of 15.3 hours per day and when active the average rate of generation has been 890 W. This may not sound too impressive for a system which has a peak output of 4 kW, but it is about what you can expect. Sunrise and sunset are times when there is naturally far less sunlight available to the panels, also we've had a lot of very overcast and rainy days.

This figure covers only April, May and half of June. I would expect that the number of hours per day of useful generation will increase in July and August, but that this will be very much less in the winter months.

The inverter gives a value of the electricity that we have generated so far as approximately 160 euros. However, we actually pay about 18.5 cents per kWh so actually we would have paid considerably more than this. The total cost of the system was about €8000, so we're currently generating at a rate which would pay back in about 8 and a half years. I expect the lower rate of generation in winter will bring this closer to the ten years that we expected. Of course, should the cost of electricity rise, this goes in our favour.

The figure for CO2 "saved" is amazing, but accurate if compared with a coal fired power station. When you burn coal you can only generate about 2 kWh per kg of CO2 produced. I did some calculations a couple of years back and found that I could easily put three or more times this much CO2 into the air each year by driving a car instead of cycling.

For the last 30 days, i.e. half of May and half of June, the average rate of generation has been higher than the average for the whole 73 days. This is of course due to May and June being brighter on average than April.

I expect that July and August will be better again, before the shorter days of winter have a less positive effect.

The effect of a shadow

This shadow doesn't look much, but its effect is surprising

Note that in the early morning the production is very much lower because our roof is aligned South West so at the start of the day the panels are entirely in shadow. As the earth rotates, the sunlight starts to fall directly, if at rather a shallow angle, onto the panels, and output increases. However, that shadow has a large effect at first.

We have 16 solar panels, wired up as two strings each of 8 panels in series. Because the panels are in series, a shadow like this across just one panel is enough to reduce the useful output of the whole string by a half. For that reason, in the morning one string has half the output of the other.

This isn't important for us as it only happens for an hour or two when the panels are least well aligned with the sun and while the output is in any case quite low. Therefore its overall importance is small.

We knew this would happen for a short period at the start of the day, and this is the reason why we didn't choose to fit two extra panels which could have been placed above the various vents on our roof. If we had placed panels there, these would have been in the shade for virtually the whole day, reducing the output of the string of which they were a part at the peak times, when the panels were best aligned with the sun, by more than the extra panels would ever provide a benefit. More panels would have brought us a lower output at a higher cost. If you're installing a system, it's very important to note where shadows are on your roof, especially at your peak hours (generally around mid-day, but later is important if you face South West and earlier is important if you face South East).

How the weather affects the output
I don't suppose many people would be surprised to learn that dark clouds reduce the output of solar panels. This can actually be quite dramatic. While the peak output of our system is 3995 W, and the system comes very close to this at some point on most cool but sunny days, when the weather is really bad, for instance when we had a hailstorm a couple of weeks back, the output drops as low as 200 W for a period.

Another thing which might surprise some people is that the peak output is not achieved on the hottest days. Once solar panels get too hot they actually generate less. On the hottest days that we've experienced, the peak has been only around 3000 W, 1/4 less than the peak on cooler sunny days. However, these were still good days for our system as they were long bright days and therefore the system stayed at a rate of 3000 W for a long time.

I should note that these effects were expected. Before fitting this system, I played around with small scale solar generation for over 20 years and I've seen the same things happen with all the other panels I've used.

Conclusion
It's working very well so far. Only hindsight can tell us if this was really worth doing, but so far it seems like not only is it a good thing for the environment, but also having paid back 2% of its cost in just a two and a half months it's a good financial investment too even though we had no subsidy to buy the system. If you can get a subsidy where you live, or even if you can't, I currently have no reason not to recommend installing a system like this.