First winter - energy input

It is almost the end of February 2026, and we’ve already had some wonderfully sunny and warm days. It makes it tempting to think that winter is over. We know better, of course — last year we still had cold days well into May. But from a solar power perspective, we can safely say: winter is behind us.

Naturally, the first question is: how did it go?
Was there enough energy?

That question can be answered in many ways. We knew in advance that we would not have enough energy during winter — and that was by design. To generate sufficient solar power on the shortest day of the year, December 21st, we would need so many panels that for the rest of the year we’d be drowning in excess energy. On top of that, such a system would simply be far too expensive.

In winter, we don’t just get less sun — we also need more energy. As temperatures drop, we try to heat the houses as much as possible using electricity. Demand goes up exactly when production goes down.

And then there’s the heat pumps. As outdoor temperatures fall, they become less efficient, requiring even more electricity to keep a house at a comfortable 20°C. That said at the moment we do not have any heat pumps installed and use the even less efficient electric heaters.

All of this inevitably leads to energy shortages in winter. So yes — there was not enough energy.

But perhaps the better question is: was the energy situation what we expected?

Since this was our first year, expectations were hard to define. We based our calculations on data from the previous year and aimed for 20 kWh per household per day for 90% of the year, and 10 kWh for 99% of the year. That means we expected around 36 winter days when we wouldn’t reach 20 kWh and would be limited to 10 kWh per day.

You might think that would be an easy question to answer. But again, it wasn’t.

Our solar panels are connected in two different ways: through our Victron system in the container, and via micro-inverters in the fields. Collecting all the data is difficult, especially since the micro-inverters occasionally lose communication, leaving gaps in the data. On top of that, when the batteries are full, solar input is shut off — energy that could have been harvested if we had a larger battery at the time, got lost.

Our setup is also far from ideal. Since the houses haven’t been built yet, we don’t have roofs for the panels. Instead, they are scattered on the ground — definitely not optimal. We also didn’t yet have our target of 35 panels per household installed. At that point, we had around 100 panels total.

To improve the situation, we ordered a 100 kWh battery system to complement the existing 10 kWh system. Unfortunately, delivery was delayed by several months, and it wasn’t operational until halfway through December.

So, let’s at least answer the question: how was the winter?

This winter in Spain was extremely wet. It made national news repeatedly, breaking record after record. For example, water reservoirs saw their largest weekly increase ever recorded — and measurements have been taken since 1988. So yes, it was wet, but also very cloudy. And clouds mean very little solar input. That part we didn’t expect. We knew we were in a cold area, but the lack of sun surprised us.

Interestingly, for a long time the highest daily solar production occurred on December 25th, and even well into mid-January — simply because that day happened to be sunny.

Overall, we had less energy than expected, but it was still manageable. Before the new battery system arrived, power would often run out in the evening. Adding more panels didn’t help, because the battery would already be full during the day. We had to carefully plan energy use, doing as much as possible while the sun was shining, and still sometimes running out before evening.

An other problem was there was only a 5kW converter, so we could not add nor use to much energy at the same time.

Once the big battery system was operational, life became much easier. Energy use timing no longer mattered as much, since the battery rarely reached its full capacity. We still had to be careful, though, so we turned the system off at 23:00 and back on at 7:00. During the day, we had enough power for lighting, computers, and some cooking. Heating was done with small gas heaters or the fireplace.

We’re not there yet. We still need:

more panels in better locations
better insulation to reduce energy demand
and ideally to start winter with pre-heated houses and fully charged batteries

Winter was challenging, but living with the sun — and sometimes without it — is part of the romance. It made the return of sunny days even more joyful. We can clearly see that the most difficult period is between November 21st and January 21st.

Next year, we’ll be doing a lot more work to improve the situation!

Is there any interesting data to look at?

Yes. I took screenshots from last winter and this winter of five panels on the container. They haven’t moved or changed in any way, which makes their energy yield directly comparable. Below are the images.

It is a bit difficult to compare but it seems there where much less days of 3kw yield per day (3kw for 5 panels will be about 20 kw for 35).