Open any manufacturer's runtime table and you'll find a line like "Refrigerator (200W): 4.1 hours." It is technically correct and practically useless, and if you size a power station around it you will buy far more battery than you need.
The figure assumes your fridge draws 200W continuously for four hours straight. It doesn't. A refrigerator's compressor runs in bursts, shuts off when the target temperature is reached, and stays off until the interior warms enough to trigger it again. The proportion of time it actually runs is called the duty cycle, and it's the single number that decides your real runtime.
What a duty cycle does to the maths
A typical household refrigerator in a normal room runs its compressor somewhere between 30% and 50% of the time. Newer, better-insulated models sit at the lower end. Older units, or any fridge in a hot garage, sit at the higher end.
That changes the arithmetic completely. A fridge labelled 200W running at a 35% duty cycle has an average draw of about 70W, not 200W. Run the same battery against 70W instead of 200W and the runtime roughly triples.
The quick version: take the manufacturer's runtime figure for your fridge's rated wattage, then multiply by somewhere between 2 and 3. That's your realistic planning range. Use 2× if the fridge is old or the room is warm, 3× if it's modern and the room is cool.
Worked example
Take a 1,152Wh station like the Bluetti AC180. Its published figure for a 200W fridge is 4.1 hours. Apply a 35% duty cycle and the average draw becomes 70W, which on the same formula gives closer to 10 hours. That is the difference between "this won't get me through the night" and "this comfortably will."
Scale that up and the effect compounds. A 3,840Wh unit like the Elite 400 shows 15.4 hours against a 200W fridge. At a realistic duty cycle you're looking at well over a day and a half on the fridge alone.
Finding your own fridge's real draw
You don't have to guess. There are three ways to get a number you can trust, in increasing order of effort and accuracy.
1. The energy label
Most refrigerators carry an annual consumption figure in kWh per year. Divide by 365 and then by 24, and you have the true average draw in watts, duty cycle already baked in. A fridge rated at 400kWh/year works out to about 46W average. That single division is more useful than any spec-sheet wattage.
2. A plug-in energy meter
An inexpensive plug-in meter left on the fridge for 24 hours gives you an exact kWh figure for a real day in your actual kitchen, including door openings and your ambient temperature. This is the number to plan a backup around.
3. Watch the cycle
If you want a rough duty cycle in ten minutes, listen. Time how long the compressor runs and how long it stays silent between cycles. Running four minutes out of every twelve is a 33% duty cycle. It's crude, but it's in the right order of magnitude.
What this means for sizing
Once you're working with average draw rather than rated wattage, sizing becomes straightforward. Multiply your average watts by the hours you need, then add 20% for inverter conversion losses.
| You want to cover | Average 50W fridge | Average 80W fridge | Sensible station size |
|---|---|---|---|
| Overnight (12 hrs) | 720Wh | 1,150Wh | 1,000–1,500Wh |
| A full day | 1,440Wh | 2,300Wh | 2,000–2,500Wh |
| Two days | 2,880Wh | 4,600Wh | 3,000Wh+ or expandable |
| Three days or more | 4,320Wh | 6,900Wh | Expandable system plus solar |
Those figures are for the fridge alone. A realistic outage also involves lights, a router, phone charging and possibly a fan, which together add roughly 100–150W of near-continuous draw — often more than the fridge itself averages.
Three things that will surprise you
The startup surge doesn't affect runtime
A compressor can momentarily pull two to four times its running wattage on startup. That matters enormously for whether your station can start the fridge at all, which is why surge ratings exist. It barely affects total energy consumed, because the surge lasts a fraction of a second. Don't let it distort your capacity maths — but do check the surge rating before buying.
A full fridge uses less energy than an empty one
Cold mass holds temperature. A packed fridge coasts between compressor cycles far longer than a sparse one. If you know an outage is coming, filling gaps with bottles of water measurably extends your runtime at no cost.
The garage is the enemy
Duty cycle is driven by the temperature difference between inside and outside the cabinet. A fridge in a 90°F garage may run at a 60% duty cycle or higher, close to doubling its average consumption versus the same unit in an air-conditioned kitchen. If your backup fridge lives in a garage, size for the worse case.
Working out which model fits?
We've mapped all ten current Bluetti models against capacity, output and cycle life — with the trade-offs each one makes.
The short version
- Manufacturer fridge runtimes assume continuous draw and understate reality by roughly two to three times.
- Find your real average draw from the annual kWh label, or measure it with a plug-in meter.
- Size on average draw, add 20% for inverter losses, then add your other loads.
- Check the surge rating separately — that decides whether the fridge starts, not how long it runs.
- A warm room or an empty fridge both push consumption up. Plan for your actual conditions.
Sizing for a fridge through an outage?
2,073Wh covers a fridge overnight with room for lights and Wi-Fi, and the cells are rated for 6,000+ cycles.
Keep reading
Your solar panels may be half wasted
Input is capped by watts, volts and amps — and watts is usually the least binding.
UPS switching speed: 0ms, 10ms, 20ms
What each tier protects, and why the smallest unit can beat the largest.
Running a CPAP from a power station
Nights per charge, and why the humidifier changes everything.
Runtime figures referenced here come from Bluetti's published tables and from the runtime formula Bluetti publishes for its own models. Duty cycle ranges are general industry figures rather than measurements of a specific appliance — your own fridge's energy label is always the better source.