How to Read an EnergyGuide Label to Size Backup Power
Figures last verified
The yellow tag on the side of your refrigerator tells you how many watt-hours that appliance uses in a day. Divide the kilowatt-hours per year printed on the label by 365, multiply by 1,000, and you have the daily watt-hour figure that every battery and solar sizing calculation starts from. The label will not tell you the appliance's surge draw, and sizing on the label alone is how people end up with a power station that trips the moment the compressor kicks on.
That is the whole method, and the rest of this page is about the parts of it that go wrong.
What the label actually is
The EnergyGuide label is a Federal Trade Commission requirement, not a manufacturer's marketing choice. The FTC's appliance labeling rule obliges sellers to attach it to refrigerators, freezers, dishwashers, clothes washers, room air conditioners, water heaters, furnaces, boilers, televisions, and a handful of other categories. You can read the rule's scope on the FTC's appliance labeling page. Notably absent from that list are microwaves, cooktops, coffee makers, space heaters, and most small kitchen appliances, which matters later.
The label carries two large numbers and a range bar. The dollar figure is an estimated yearly operating cost, calculated against a national average electricity rate that is usually out of date and almost never matches your utility. Ignore it. The number you want is the smaller one, usually printed underneath or beside it: estimated yearly electricity use, expressed in kilowatt-hours per year.
That figure comes from a standardized Department of Energy test procedure. The appliance is run in a controlled chamber at a fixed ambient temperature with the doors closed, and the resulting consumption is extrapolated to a year. It is a real measurement of a real unit under real electrical load. It is also a laboratory measurement, which is the first place your sizing will drift.
Turning the label into watt-hours per day
Take a common example. A full-size refrigerator labeled at 400 kWh per year works out to roughly 1.1 kWh per day, which is 1,100 watt-hours.
That single number does more work than anything else in backup power planning. A power station's capacity is printed in watt-hours. A battery bank's capacity is amp-hours times nominal voltage, which is also watt-hours. Solar production is watt-hours per day. Once the appliance is expressed in watt-hours per day, everything is in the same unit and the arithmetic becomes trivial.
If the label is missing, gone, or unreadable, the manufacturer's model page usually lists the same figure, and the Energy Star product finder carries it for certified models. Failing both, the nameplate inside the door gives volts and amps, but that product is the maximum continuous draw, not the daily average, and using it will oversize your system by a factor of five or more.
The derating nobody applies
Here is where sizing goes wrong in the direction that costs you money rather than the direction that leaves you cold.
The 1,100 watt-hours from the label is what the appliance consumes at the wall under test conditions. What your battery has to supply is larger, for three compounding reasons.
Inverter losses come first. A power station stores DC and your refrigerator wants AC, so an inverter sits between them, and inverters run at roughly 85 to 90 percent efficiency under moderate load. Efficiency falls off at very light loads, which is exactly the condition a refrigerator creates for most of every hour. Call it 85 percent and you have turned 1,100 watt-hours of appliance demand into about 1,300 watt-hours of battery draw.
Usable capacity comes second. A station advertised at 1,000 watt-hours does not deliver 1,000 watt-hours to your appliance. Some of that is the inverter loss already counted, and some is the battery management system holding back the bottom of the pack to protect cell life. Lithium iron phosphate chemistry is generous here compared to lead acid, but the advertised figure is still the gross capacity of the cells, not what comes out of the AC outlet.
Ambient temperature comes third, and it is the one people are most surprised by. The DOE test runs at a controlled room temperature. A refrigerator in a house with no air conditioning during a summer outage works considerably harder than the same unit in a lab, because the temperature differential it is fighting is larger and the door is being opened by people checking whether the food is still cold. Consumption in that scenario runs well above label. So does a chest freezer in an unconditioned garage.
Stack the three and the practical planning figure for a 400 kWh-per-year refrigerator during a warm-weather outage is closer to 1,500 to 1,800 watt-hours per day than the 1,100 the label implies. That is the number to carry into the solar sizing calculator, not the raw label figure.
What the label will never tell you
The label reports average consumption over time. It says nothing about instantaneous demand, and instantaneous demand is what determines whether your inverter can start the appliance at all.
Anything with a compressor or an induction motor draws a large current spike at startup, lasting a fraction of a second to a couple of seconds, while the motor overcomes inertia and comes up to speed. Refrigerators, freezers, well pumps, sump pumps, air conditioners, and power tools all do this. The spike can be several times the running draw. A refrigerator that averages 40 to 60 watts across an hour of duty cycling might pull 300 to 800 watts while running and briefly spike past 1,000.
None of that appears on the yellow tag. What you can find is the nameplate inside the appliance, which sometimes lists locked rotor amps, abbreviated LRA. Multiply LRA by your line voltage and you have a rough worst-case surge in volt-amps. If the nameplate gives only full load amps, or FLA, that is the running figure and the surge sits above it.
This is the gap the appliance surge and running watts table exists to fill, and it is the reason a system sized purely on daily watt-hours can be simultaneously large enough to run an appliance for a day and unable to turn it on.
Two numbers, two different jobs
Sizing backup power means answering two independent questions, and the label answers only one of them.
The capacity question asks how long. It is answered in watt-hours, it comes from the label, and it determines how big a battery or how many panels you need. Get it wrong and your power runs out earlier than planned.
The output question asks whether at all. It is answered in watts, both continuous and surge, it comes from the nameplate or from measurement, and it determines whether the inverter can carry the load. Get it wrong and nothing runs regardless of how large the battery is.
Confusing the two is the single most common sizing error in this category, largely because manufacturers name their products after the watt figure and sell them on the watt-hour figure. That distinction is worked through in watts versus watt-hours.
Doing it for a whole house
Repeat the label reading for every appliance you intend to keep running, sum the daily watt-hours, and apply the derating once at the end rather than to each item.
Be honest about the list. A refrigerator, a chest freezer, a few LED lights, a phone charger, and a router is a realistic critical load and lands somewhere around 2,000 to 3,000 watt-hours per day for most households. Adding a window air conditioner or an electric water heater changes the answer by an order of magnitude, and adding an electric range or a central air system moves it outside the range of any portable equipment. Which loads are genuinely out of reach is covered in what a portable power station cannot run.
If you are sizing around a generator rather than a battery, the daily watt-hour figure matters less and the continuous and surge watt figures matter more, which is why generator sizing works from a different starting point than this page does.
Verify with a meter if the stakes are high
A plug-in energy monitor costs less than a tank of generator fuel and removes all of the estimation above. Leave one on the refrigerator for a week, read the accumulated kilowatt-hours, and divide by seven. That gives you your appliance, in your house, at your ambient temperature, with your door-opening habits, which no label can.
For anything hardwired, or for a whole-panel measurement, a clamp meter on the service conductors or a circuit-level monitor does the same job at higher cost.
The label is the fast estimate. The meter is the real number. For a single refrigerator the difference rarely changes which product you buy, but for a multi-thousand-dollar system it routinely does.