A battery is sized by the evening, not by the array

The wrong starting number

Sizing conversations tend to begin with the array. So many kilowatts of panels, therefore so many kilowatt-hours of battery, as though the two scale together. They don’t, because the battery’s job is not to hold everything the array can make. Its job is to carry the household from the point generation stops to the point it resumes.

That makes overnight consumption the governing figure. Everything else is secondary: array size, roof orientation, even annual generation. If a house draws a certain amount of electricity between sunset and sunrise, a battery larger than that quantity has capacity that will be full at dusk and still partly full at dawn, doing nothing.

Finding the number is not hard where half-hourly data is available. Take the consumption between dusk and dawn on a set of ordinary days, in more than one season, and look at the spread rather than the average. That range is the design input.

Two loads, and only one of them is visible

Overnight consumption splits into two parts that behave very differently. There is the baseline: the refrigerator cycling, the router, the alarm, standby draws, anything with a clock. It runs at a low and remarkably steady rate for the whole night, and multiplied by the hours it accounts for more than most households expect.

Then there is the evening peak, which is cooking, hot water, laundry, television, lights, all compressed into a few hours. It is much larger in rate and much shorter in duration, and it is the part people picture when they imagine a battery in use.

The two impose different requirements. The evening peak tests the battery’s power rating, meaning how many kilowatts it can deliver at once, which is a separate specification from its capacity. The overnight baseline tests capacity. A battery can be large in kilowatt-hours and unable to run an electric shower, because those are different numbers on the datasheet.

Winter and summer want different sizes

The uncomfortable part of battery sizing is that no single capacity is right in both seasons. In midsummer there is a large surplus and long days, so a bigger battery has plenty to absorb and something to do. In midwinter the array may not generate enough in a whole day to fill even a modest battery, and the extra capacity is inert.

Which means a battery sized for summer surplus is oversized for the season when imported electricity matters most, and a battery sized for winter charging is too small to catch the summer excess. There is no configuration that resolves that, only a choice about which season to optimise for.

The usual resolution is to size for the more common case rather than the extreme, and accept that the battery will be undersized on the sunniest days and underused on the darkest. That is not a failure of the equipment. It is the seasonal shape of solar meeting a fixed box, and any honest sizing exercise ends there.

Usable capacity, and how quickly it shrinks

The capacity on the box is not the capacity in service. Manufacturers reserve a portion at each end of the range for the health of the cells, so the usable figure is smaller than the nameplate and should be stated separately. Round-trip losses then take another share of everything that passes through.

On top of that, most systems hold back a reserve if backup during an outage is configured, since a battery that discharges to empty every evening has nothing left when the supply fails. That reserve is genuinely useful and it is capacity removed from daily cycling.

Work backwards through those deductions and a battery advertised at one figure delivers noticeably less to the sockets. It is not deception, provided the numbers are disclosed, but it does mean that comparing nameplate figures between products compares the wrong quantity.

Bigger is not safer, and it is often worse

The instinct when uncertain is to buy the larger unit, and here that instinct usually costs money. A battery earns by cycling: each full charge and discharge displaces imported units, and its value accumulates through repetition. Capacity that is never emptied is capacity that never earns, while still ageing on the calendar.

There is a second penalty. A larger system generally has larger standing losses, since more cells, more monitoring electronics and more conversion hardware all draw a little continuously. Those losses apply whether the capacity is used or not.

The defensible reasons to oversize are specific rather than general: a car to charge, a heat pump arriving, a tariff with a narrow cheap window that has to be filled quickly, or a genuine requirement for extended backup. Absent one of those, a battery that empties most evenings and refills most days is working properly, and one that hovers half full is a purchase that was larger than the household.