Power Station Sizing Calculator

Why most sizing advice buys you too much battery

Pick any wattage chart and it will tell you a refrigerator is around 725 watts. Multiply that by twenty-four hours and you get 17,400 watt-hours a day, which is more than ten times what a fridge really uses and points you at a battery costing thousands.

The error is in the chart's own small print. The Department of Energy's figures are explicit that the number stamped on an appliance is the maximum it can draw, and that to estimate a refrigerator you should divide the hours it is plugged in by three, because a compressor runs on a thermostat rather than a clock. A fridge is idle most of the time.

There is a better number still, and it is already on your fridge. The yellow EnergyGuide label gives kilowatt-hours per year, measured under a standard test that has the duty cycle built into it. A middling full-size unit is around 500 kWh a year, which is 1,370 watt-hours a day. Across certified models the figure runs from roughly 350 to 800 kWh a year, so walking over and reading your own label is the single most useful minute you can spend on this.

This calculator uses the annual label figure for anything that cycles and watts-times-hours for everything that does not, because those are two different kinds of load and treating them the same is what produces the ten-fold error.

Watt-hours and watts are different questions

A battery has two limits and they constrain different things. Capacity, in watt-hours, decides how long it lasts. The inverter, in watts, decides what you can plug in at all. A unit with an enormous battery and a small inverter will run a fridge for two days and still refuse to boil a kettle.

So the calculator reports both. The continuous figure assumes everything you ticked is drawing at once, which is pessimistic and is meant to be: it is the number that decides whether the evening everyone is home goes smoothly.

Add one surge, not all of them

Anything with a motor — a compressor, a well pump, a furnace blower — draws several times its running power for a fraction of a second as it breaks away. That spike is what trips an inverter.

The mistake is adding every appliance's surge together. Motors start when their own thermostat or pressure switch tells them to, and the odds of two breaking away in the same instant are slim, so the convention is to take the running total and add only the largest single surge on top. Adding them all gives a figure two or three times too large, which is how people end up with a unit they did not need.

One caveat we would rather state than hide: the surge multipliers used here are sizing conventions, not measurements. The honest figure for your compressor is the locked-rotor current on its own nameplate.

The recharge question decides everything past day one

A power station is a tank, not a well. Sized for a day, it is excellent; sized for a week with no way to refill it, it is a day of comfort followed by six days of a heavy box.

The solar estimate here is deliberately pessimistic. Panel ratings are measured in a laboratory at full sun and 25 degrees, so we assume three quarters of the rating and four hours of usable sun. A portable panel flat on the ground in December will do considerably worse, and a day of rain does nothing at all.

If an outage lasting more than a couple of days is the thing you are planning for, read when the lights stay off, which deals with that case directly, and the power outage emergency kit for what sits alongside the battery.

What this calculator assumes

Every appliance figure is typical rather than measured, and each row names where it came from in the table below. Inverter and conversion losses are taken as 15 percent, which is why the capacity to shop for is higher than the energy you need. Solar assumes four hours of usable sun at 75 percent of panel rating. Depth of discharge is not modelled separately; on the lithium iron phosphate chemistry these units use, treating the published capacity as usable after the inverter loss is close enough for choosing between models.

Nothing here is a measurement of your home. It is arithmetic applied to typical values, and its purpose is to get you to the right order of magnitude before you spend money, then to tell you which three labels are worth reading to do better.