Summer Solstice 2022: Solar PV update

The Summer Solstice seems like a good a time to take a look at the first half-year of generation from our 12 solar panels and the effect of our Tesla Powerwall battery.

Last year at around this time I wrote that – having been off-grid for 90 days – I felt like I was ‘floating’.  And then just a couple of days later I came back down to Earth after several consecutive days of unseasonably dull weather led to me have to buy some full-price electricity! In mid-summer!

This year we have been off-grid for only 50 days so far and I will discuss the reason for the difference below.

In case you don’t have time to read the article, it’s performing pretty much as expected and very similarly to last year.

Let’s begin.

Solar Generation.

One of the difficulties in communicating data from a solar PV system is its variability. The data obviously change seasonally, but also on daily, weekly and monthly time scales. So I have chosen to present the same data in several ways.

Let’s start with a basic chart showing the average daily generation (in kWh) over the last two years.

Click image for a larger version. Average daily solar generation (kWh) for each month of this year and last year.

This chart shows that generation this year is generally a little better than last year, with the exception of last May, which was enchantingly sunny.

Now let’s look at the daily data, and different averages.

Click image for a larger version: you’ll need to do this to see anything! This chart shows the daily generation this year, and the 5-day running average  of generation this year and last year. See the text for details of the other data on the graph.

The graph above is complicated, showing how various quantities (expressed as kWh/day) change versus day of the year.

Primarily it shows daily solar generation as a light green line. Notice the day-to-day variation: even in June, daily generation can fall to 3.8 kW/day even when the average generation is ~15 kWh/day.

Also shown are the 5-day averages of solar generation this year and last year. The 5-day average shows the smoothing effect of the use of a 13.5 kWh battery. When the 5-day average falls below demand, then it’s likely we will need to buy some electricity from the grid. You can see that this happened in mid-summer last year.

Notice that the 5-day averages show peaks and troughs that can last for weeks.

My expectations for the system are shown as a dotted green line (just a simple mathematical guess)  and as yellow monthly data points estimated using a multi-year European database. Generation is broadly in line with expectations.

The graph also shows nominal household demand as two red-dotted lines.

  • The horizontal line (10 kWh/day) corresponds to daily demand throughout the year.
  • The demand peaking in winter represents the electricity used by the heat pump to heat the house. This demand has only been present this winter – previously, heating was supplied by a gas boiler.

So the generation has interesting day-to-day variability, but when the 5-day average exceeds the average demand, then with the aid of the battery, we stand a good chance of being able to be off-grid for a sustained period.

Looking at the graph, the heating demand was still non-zero in April – and this delayed the point at which we were able to go off-grid.

The graph below shows actual (rather than nominal)  domestic use of electricity and the electricity drawn from the grid.

Click image for a larger version: This chart shows the ±7-day averages of the electricity we use in the house for heating and the electricity we draw from the grid. The difference between the two curves is supplied by the solar PV/battery system. Periods when the house has been off-grid are highlighted in red.

Another way of looking at the data.

Plotting day-to-day generation tends to emphasise the variability of the data.

Plotting cumulative generation through the year tends to de-emphasise the variability and highlight the similarity of generation from one year to the next.

The graphs below show cumulative generation versus day-of-the-year for this year and last year. The upper graph shows the whole year and the lower graph the first half of the year only.

Also shown as dotted lines are the MCS suggestion for likely annual generation and an empirical curve based on the MCS figures that I use to guide my expectations.

Click image for a larger version: This chart shows the cumulative generation this year and last year. See the text for details of the other data on the graph. Broadly speaking, the data are very similar.

Click image for a larger version: This chart shows the cumulative generation for the first 6-months of  this year and last year. See the text for details of the other data on the graph. Broadly speaking, the data are very similar. The very dull period in midsummer last June can be seen as a flat portion on the 2021 generation curve.

Remember that year-to-year variability in generation is typically ±5% (link), so that yearly variations of ~200 kWh on an expected generation of 3,780 kWh is quite normal.


Overall, everything is proceeding as expected. The combination of 12 x 340 Watt Q-Cells panels and the 13.5 kWh Tesla Powerwall battery has been astonishing.

  • We save money in summer – because we are off-grid.
  • We save money in spring and autumn because solar PV is still significant.
  • We save money in winter because we can buy cheap-rate electricity and use it during the day.

The reality of running the house – with fridges and freezers and computers and washing machines and air-conditioning and even tumble dryers – solely from the sunshine for several months a year, still warms my heart.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: