How much battery do you need for 800 W of balcony solar?

There are two ways to size storage, and they answer different questions. Starting from the house asks "how much do I need to keep running". Starting from the array asks "how much of what I generate can I hold". This page does the second one, because the question that brings people here is about a specific set of panels they already have or are about to buy.

The arithmetic is short. Daily generation is panel wattage multiplied by peak sun hours, then multiplied by system efficiency. Peak sun hours are not daylight hours: they are the hours at a thousand watts per square meter, so a long summer day in a cloudy place can still be four effective hours. System efficiency covers inverter conversion, wiring loss, soiling and heat, and 80 % is a reasonable planning figure for a small system.

An 800 W array in a 4.5 hour location therefore makes 800 x 4.5 x 0.8, or 2,880 Wh a day. That is the number the battery has to hold. Nothing about your appliances changes it, and nothing about the brand of the battery changes it either.

If instead you know your daily consumption and want to work backward, the storage sizing calculator starts from the load side and reaches a nameplate figure the other way. The two numbers agree when the array is correctly matched to the load, and diverge when it is not, which is itself a useful check.

Sun hours vary more than panel wattage. The same array makes almost half again as much in a good month as in a poor one, which is why a battery sized in July runs dry in December. The table below uses 80 % efficiency throughout and three realistic sun-hour figures.

Read the 800 W row down and the point is hard to miss: the same hardware swings from 2,240 to 3,520 Wh depending on where and when. A battery sized to the middle figure will be full and clipping in June and empty by mid-afternoon in January. If the array will be used year round, size the battery to the winter row and accept that it fills quickly in summer.

The 1,600 W row is included because two 800 W strings on one balcony is a common upgrade, and because it makes the next section unavoidable: doubling the array does nothing for you if the charger cannot accept the current.

Every portable station and most hybrid inverters have a maximum solar input. It is printed in the specification, it is almost never in the marketing copy, and it silently decides whether your array is useful or decorative. A station that accepts 400 W of solar will draw 400 W from an 800 W array on a clear day and discard the rest.

That is not a defect. The charger is a power converter with its own rating, and pushing more current through it than it was designed for would damage it. What it does mean is that array and charger have to be sized together, and that "buy the biggest panel" is bad advice unless the input can take it.

The calculator in the right-hand panel reports this as a capture percentage: the share of the array's nameplate the charger can actually accept. Below 100 % you are carrying panels that only contribute on cloudy days, when output falls under the cap. That can still be worth doing, but it should be a deliberate choice rather than a surprise.

If you are choosing panels and an inverter as a set, the plug-in solar panel comparison lists what is sold in the US and how the certification claims differ. If you are still deciding how many panels you need at all, the panel count calculator works that out from your own sun hours.

Autonomy is how many days the battery can hold without any sun. One day covers the gap between evening and the next morning, which is the minimum for a grid-tied array that produces every day. Two and three days are for weather, not for routine operation.

The single boxes we track stop at 4,096 Wh, so a two-day buffer for an 800 W array already leaves the portable category. That is the point where the honest options change: an expansion battery on a line that offers one, two units in parallel where the maker supports it, or a fixed hybrid inverter with a rack battery.

The jump from one day to two is not twice the cost for twice the value. It is the difference between a system that runs out whenever tomorrow is cloudy and one that does not, and that is usually worth more than the extra capacity costs.

Four limits do not move no matter how much capacity you buy, and knowing them prevents most bad purchases.

It cannot make the sun come out. Three days of autonomy means three days, not three weeks. A long storm exhausts any battery you can afford, and the only thing that changes that is a generator or a grid connection.

It cannot recover a shaded panel. A single panel in shadow can drag a whole string down. Storage stores what it is given, so a badly sited array with a large battery is just an expensive way to store very little.

It cannot help with a load it cannot start. Capacity and power are separate ratings. A large battery with a small inverter still cannot start a well pump, and the fridge runtime calculator shows the same split for compressors.

It does not raise the value of surplus. Any production beyond what you store and use is either exported for a credit or lost. In places without a net metering credit, an oversized array plus an oversized battery is two ways of losing money at once.

Charging time is the nameplate divided by the rate at which energy actually arrives. If the array and the charger can both handle 800 W, a 3,388 Wh battery takes about 4.2 hours of full output, which fits inside a good day. If the charger accepts only 400 W, the same battery takes about 8.5 hours of full output, which almost no winter day provides.

That is the arithmetic behind a complaint that shows up constantly in reviews: the battery never reaches full charge. The unit is not faulty. The array is bigger than the charger, so the system only ever runs at the charger's rate, and the day is shorter than the required hours.

Two conclusions follow. First, size the array to the charger, not past it, unless you deliberately want the extra panels for cloudy days. Second, choose a station whose input is at least half your array's nameplate, because real output rarely exceeds half of nameplate except around noon on a cold clear day.

The figure below shows the same 2,880 Wh a day production against three autonomy targets, which is the visual version of the table above.

All fourteen units below are in the current US market, with the capacity and input figures that matter here. Capacity is nameplate, so multiply by 0.85 for usable; the last column is the share of an 800 W array the unit can accept, which is the number that decides whether the panels are worth owning. For a wider comparison across situations, see the portable power station ranking.

The shape of that list is the point. An 800 W array produces more in a morning than most of these units hold, which is why the smallest four are load-shifters rather than storage. Only the last one covers a full day, and none of them covers two — that requires expansion batteries or a different class of hardware.

Buying the battery before checking the input. The charger's solar input decides how much of the array you can use. A 3 kWh battery behind a 300 W input fills slowly no matter how big the array is.

Sizing to summer. December production can be half of June in the same location. A battery sized on a July figure will be empty all winter, and winter is when the grid fails most often.

Assuming usable equals nameplate. Depth of discharge and inverter losses take about a quarter before the first appliance is switched on. Planning on the printed number is how people come up an hour short.

Storing more than you generate. A 10 kWh battery behind an 800 W array is empty most of the year and never full. Match storage to production, then extend both together.

Buying storage for savings instead of backup. Without a time-of-use rate or an export credit, storage rarely pays back on its own. It pays back as outage insurance, which is a different budget and a different set of numbers.

If what you actually need is to know how long a given battery runs a specific appliance, the fridge runtime calculator works that through. If you have not yet fixed the array size, start with the panel count calculator and come back with a wattage.

If you are wiring a battery bank rather than buying an all-in-one box, the controller rating has to be matched to the array as well: the charge controller sizing calculator turns array watts and bank voltage into the amp rating you need.

If the panels are already on the wall rather than planned, the guide on adding a battery to plug-in solar covers the three architectures and why a retrofit is usually not one of them.

Panel output depends on location: the sun hours table lists peak sun hours, winter and summer, for one city in every state.

For the recharge side of the same question — whether the panels can refill the bank day after day — see the comparison of solar generators for home backup.

The charger between the panels and the bank has its own limit; size it with the charge controller calculator.

Once the bank is sized, the cost comparison covers what each route charges for it.

For what the pack chemistry changes and what it does not, see LiFePO4 versus NMC.

If the bank has to cover a multi-day outage rather than one night, the outage capacity calculator adds the days and the array.

Every figure here can be recomputed by hand: daily generation is array watts times sun hours times efficiency, nameplate is usable divided by depth of discharge, and charge time is nameplate divided by the smaller of array and charger. When a value on this page changes, the changelog records it. The way corrections are published is described on the disclosure page, and the privacy page covers what this site collects.