Seasonal dispatch: a battery sized for summer is the wrong tool in December

This article was generated with AI assistance from cited sources and has not been individually reviewed by an editor.

A 1 kWp rooftop array in Brussels produces about 126.6 kWh in May and 33.4 kWh in December — a 3.8× gap. A home battery sized for the sunny half of the year spends the dark half doing almost nothing. Most systems are still sold as if the two seasons were the same, and that single assumption is quietly deciding whether the battery ever pays for itself.

Solar output across Northern Europe is not a smooth curve. It is a steep seasonal wave, and the winter trough is deeper than most installers put on a quote. We pulled the monthly production figures for a 1 kWp crystalline-silicon array in four European capitals from the European Commission’s PVGIS tool (JRC, accessed 17 August 2026). The spread is not subtle.

kWh produced per 1 kWp per month, fixed 35° south-facing array, 14% system loss, crystalline silicon. Source: JRC PVGIS (SARAH2 radiation data, 2005–2020), accessed 17 August 2026.

Three things in that table matter for anyone sizing a battery.

First, the best month to worst month ratio. In Brussels it is 3.8× (126.6 vs 33.4); in Amsterdam 4.8× (132.2 vs 27.8); in Berlin 4.5× (132.0 vs 29.1). Only Paris, further south, softens it to 2.9× (134.6 vs 45.9). That ratio is the whole problem: a battery that can absorb a full May afternoon has four to five times more energy available than it will ever see in December.

Second, the winter quarter is genuinely small. Across December, January and February, the Brussels array produces about 123.7 kWh per kWp — roughly a third of the 362.4 kWh it produces across June, July and August. There is no sizing trick that fixes this. In winter there is simply not enough midday surplus to charge a battery most days.

Third, the annual figure hides all of it. An installer can quote “1,039 kWh per kWp per year” and be completely accurate, while the customer hears “steady production”. It is not steady. Most of it lands between April and September.

Two jobs, one box

A home battery does two different jobs in two different seasons, and they do not require the same capacity.

In the summer half, the job is time-shifting. The roof overproduces from roughly 10:00 to 16:00, and the household consumes most in the evening. A battery stores the midday surplus and releases it after sunset, lifting self-consumption and cutting what you buy from the grid. This is the job every quote describes, and it is the one that drives sizing upward — because more capacity means more of that midday surplus is captured.

In the winter half, the job is nearly gone. There is little surplus to store, so the battery sits mostly empty. Its remaining value is narrow: it can still flatten a short evening peak to shave a capacity-tariff charge in Flanders, or time-shift cheap night or dynamic-tariff hours into the evening. But on a plain flat tariff with no capacity charge, a winter battery is largely a very expensive paperweight for three to four months of the year.

What the seasonality does to payback

Payback is not earned in the summer. It is earned across the whole year, and the winter drag is where the numbers usually fall apart. Take a Flemish household on the 2026 capacity tariff, which charges an average of €53.39 per kW of monthly peak per year, before VAT (VREG). Belgian households in the standard band paid about €0.3499 per kWh in the second half of 2025 (Eurostat, data extracted April 2026). Both prices are real; the question is how many months a year a battery actually reduces them.

Seasonal dispatch model; energy price and capacity-tariff rate from VREG and Eurostat as cited above.

The uncomfortable conclusion: a battery that only pays back through summer self-consumption needs a very long summer to justify its installed price. A battery that also earns in winter — by shaving the capacity-tariff peak, or by cycling on a dynamic tariff that pays for cheap-hours charging — shortens that payback materially. In a country with no capacity charge and a flat tariff, the winter months contribute almost nothing, and the honest payback model should show that as a near-zero winter line rather than burying it in an annual average.

Sizing and setting the battery for the season

The fix is not a bigger battery. It is a battery sized for the job that actually exists, plus a seasonal dispatch setting that changes what the box is allowed to do.

• Size to the summer evening, not the summer noon. The capacity you need is set by how much surplus you can realistically use between sunset and midnight — not by how much the roof can produce at 13:00. Sizing to the noon peak is how households end up with an oversized unit that is never more than half-full even in July.
• In winter, set the battery to peak-shaving, not self-consumption. Where a capacity tariff applies (as in Flanders), reserving a small reserve and discharging it against the household’s evening peak is worth more in winter than chasing a surplus that is not there. The VREG capacity tariff makes the winter peak the one thing the battery can still reliably monetise.
• On a dynamic or time-of-use tariff, let winter charging come from the grid. When overnight or midday prices dip below the evening price, the battery earns by buying cheap and selling to the household at peak — a winter revenue stream that does not depend on sunshine at all. This is the only way a battery in a flat, cloudy January keeps earning.
• Re-run the payback model month by month, not as one annual average. A model that shows zero winter contribution and still pays back is honest. A model that smears the summer saving across twelve months is not.

The action list

• Ask the installer for the monthly production curve of your roof (PVGIS or a site-specific tool), not the annual total.
• Ask them to model the battery against your evening consumption in each of the four seasons, not against daytime production.
• If you are in a capacity-tariff region, get the winter peak-shaving saving as a separate line item.
• If you are on — or can move to — a dynamic tariff, ask for the winter grid-charging scenario, and what it adds to payback.
• After the first winter, check the model against reality. The first full winter is where an honest sizing shows up.

Sources

• European Commission Joint Research Centre (JRC) — PVGIS, monthly photovoltaic output for 1 kWp arrays in Brussels, Amsterdam, Berlin and Paris (SARAH2, 2005–2020): link (accessed 17 August 2026)
• VREG (Flemish energy regulator) — capacity tariff, 2026 rate of €53.39 per kW of monthly peak per year (excl. VAT): link (accessed 17 August 2026)
• Eurostat — Electricity price statistics, Belgium household price €0.3499/kWh (H2 2025, data extracted April 2026): link (accessed 17 August 2026)

Method and sources as noted inline. No manufacturer or installer reviewed this before publication. Monthly figures are modelled production from PVGIS, not measured output from any single household.