A customer called last month, upset. He had bought a 5kWh battery in the spring, ran it through one cloudy stretch, and watched it die by breakfast. He wanted to know if the battery was defective. It was not. The battery was fine. It was just half the size his house needed, and nobody had done the math before he paid.
This is the most common phone call we get. Somebody buys a battery based on a number that sounded big, then finds out the hard way that "big" and "enough" are different words. So let's do it the right way, in order, with real numbers you can copy for your own place.
There are three steps. Load survey, days of autonomy, voltage class. Skip any one of them and you are guessing.
Step one: the load survey
A load survey is a list of what you actually want to run when the grid is down, and how much energy each thing eats in a day. Watts and watt-hours are different animals. Watts tell you how hard something pulls at one moment. Watt-hours (and kilowatt-hours, which are just 1,000 of them) tell you how much it uses over a whole day. Battery size is a watt-hour question, so that is the number we chase.
You do not need to back up the whole house. You need to back up the stuff that matters when the power is out. For most homes on the Gulf Coast that short list looks like this:
- Refrigerator: about 1.5kWh a day
- Chest freezer: about 1kWh a day
- Lights, phones, router, a laptop: about 0.5kWh a day
- Well pump or a couple of fans: about 0.5kWh a day
Add those up and you land near 4kWh a day. Your house will be different. A medical device, a second freezer full of game, a window AC unit for the bedroom, all of that changes the total. The point is you write down every load, put a daily watt-hour number next to it, and add. That sum is the foundation. Everything else builds on it.
One honest note. If you want to run central air off a battery through an outage, the number stops being 4kWh and starts being 30 or 40. Central AC is the single biggest load in most houses. Back up the bedroom with a mini-split or a window unit instead, and the math stays sane.
Step two: days of autonomy
Days of autonomy is how long the bank carries your loads with no help from the sun or the grid. This is where people either lie to themselves or overspend, so be honest.
If your panels are on the roof and a normal sunny afternoon refills the bank, one day of autonomy is often plenty. The battery is really just a bridge from sunset to sunrise. But a hurricane does not leave sunny afternoons behind it. It leaves three gray days and a wet sky. For a Florida house that wants to ride out a real storm, two days is the honest number for most homes. One day is optimistic. Five days is buying battery you will almost never use.
So take the 4kWh a day from step one and multiply by two days. That is 8kWh of energy you need to pull out of the bank.
Now one adjustment. You never drain a battery to zero. We only sell LiFePO4, and we plan around using about 80% of the nameplate rating to keep the cells healthy and give yourself a margin. So divide that 8kWh by 0.8, and you get 10kWh of battery you actually need to buy.
That is the whole calculation. 4kWh a day, times 2 days, divided by 0.8, equals a 10kWh bank.
Step three: voltage class
Last choice, and it is quick. Battery banks come in 12V, 24V, and 48V flavors. For a house, 48V wins almost every time.
Here is why in one sentence: higher voltage means lower current for the same power, and lower current means thinner wire, smaller fuses, less heat, and less loss. A 10kWh bank at 12V has to shove around 830 amps to make big power. That same bank at 48V moves a quarter of the current for the same watts. Thinner cable, cheaper lugs, cooler runs.
Twelve volt has its place. A van, a small cabin, one appliance. But a whole-home build at 12V is fighting physics and paying for the privilege in fat copper. Pick 48V and move on.
For the record, that 10kWh at 48V works out to roughly a 200Ah battery, which is one larger wall unit or a pair of common 5kWh modules stacked together. You do not have to do that conversion by hand. The point is that once you know the kWh and the voltage class, the shopping list writes itself.
The two mistakes that make people buy twice
First, sizing to the sticker instead of to the load. The battery that sounds impressive in the ad is not automatically the battery your house needs. Do the survey first, then shop.
Second, being dishonest about days of autonomy. One day feels cheaper today. Then the first multi-day outage arrives and you are sitting in the dark at 6 a.m. deciding which is defective, the battery or the plan. It was the plan.
Run the three steps and you will land on a real number instead of a hopeful one. If you want a hand with your load survey, walk through the build at apesolar.com/build, or call the shop and we will run the math with you before you spend a dollar.
Ready for Anything.