Portable power station for a refrigerator: tested run times

By Watt Power Station · Updated

A refrigerator plugged into a portable power station in a dark kitchen during a power outage

In short

  • Longest run time we track: Jackery HomePower 3600 Plus, 38.58 hours (1.6 days).
  • Lowest price per hour of fridge time: Dabbsson 2000L. Its $649 store price works out at $30.90 for each hour it ran the fridge.
  • Least battery used per hour: Anker SOLIX C2000 Gen 2, 66 Wh for each hour the fridge ran.
  • Shortest: Oscal PowerMax 2400, 12.58 hours. It used 55.41% of its energy on itself.
  • Size for a full day: about 1,600 Wh to 2,200 Wh of usable capacity, for a fridge like CNET's.

A large portable power station can keep a full-size refrigerator running for about a day. In CNET's lab test, the 10 models we track ran the same fridge for 12.58 to 38.58 hours. Battery size is only part of the story. The power a station spends on itself decides how much reaches the fridge.

How long will a portable power station run a refrigerator?

About a day for most large units: 8 of the 10 we track lasted 20 hours or more. CNET ran 19 large and extra-large power stations on one fridge until each was empty. The fridge was a full-size Whirlpool refrigerator (118 V, 60 Hz, rated 7.2 A at full load). CNET published the results on Jul 23, 2026. Here are the models we track, longest first.

Fridge run time for each portable power station we track, from CNET's test
Model Battery (rated) Fridge run time Used on itself Price per fridge hour
Jackery HomePower 3600 Plus 3,584 Wh 38.58 h (1.6 days) Not published $44.30 ($1,709)
Jackery HomePower 3000 3,072 Wh 32.05 h (1.3 days) 36.4% $42.09 ($1,349)
Anker SOLIX F3000 3,072 Wh 28.97 h (1.2 days) 37.81% $48.33 ($1,399.99)
Pecron E3600LFP 3,072 Wh 28.15 h (1.2 days) 15.75% No price checked
Anker SOLIX C2000 Gen 2 2,048 Wh 27.40 h (1.1 days) 18.21% $32.85 ($899.99)
BLUETTI Elite 200 V2 2,074 Wh 26.32 h (1.1 days) Not published $34.16 ($899)
BLUETTI FridgePower 2,016 Wh 23.43 h (1.0 days) 20.24% $38.37 ($899)
Dabbsson 2000L 2,048 Wh 21.00 h (0.9 days) Not published $30.90 ($649)
Goal Zero Yeti 1500 (6th gen) 1,505 Wh 16.50 h (0.7 days) Not published $90.91 ($1,499.95)
Oscal PowerMax 2400 1,872 Wh 12.58 h (0.5 days) 55.41% $51.59 ($649)

"Used on itself" is CNET's figure for the share of the battery that went on keeping the station's own inverter and electronics running. "Price per fridge hour" is the brand-store price we checked, divided by the run time.

CNET also tested 9 units we don't track yet. The longest of all 19 was the Fossibot 3600 Pro at 44.65 hours (1.9 days).

Why doesn't a bigger battery always last longer?

Because the power station uses power too. A fridge's compressor switches on and off, so for much of the test the station feeds a small load. Its inverter stays on the whole time, and that costs energy every hour.

From CNET's figures we can split each battery's energy between the fridge and the station (7 units):

Our light-load data shows the same effect. At a 34 W load, Outdoor Life got 49% to 71% of rated capacity from the 6 units we track that it tested. At the heavier loads we count, the models we track deliver 75% to 96%. How we count tests

What size power station do I need for a refrigerator?

Size it on usable capacity, not the number on the box. In CNET's test, 7 of the 8 models we track with a CNET capacity figure used 66 Wh to 91 Wh for each hour the fridge ran. The Oscal PowerMax 2400 used 134 Wh. For a fridge like CNET's, that means:

Usable capacity is what a unit delivers in independent tests. Every review page shows it, and the real capacity table lists all of them. Your fridge may use more or less than CNET's, so the next section shows how to work it out.

How do I work out the run time for my own fridge?

Three steps, with a worked example:

  1. Find your fridge's average watts. Take the kWh per year from its yellow EnergyGuide label or its manual, and divide by 8.76. For example, 500 kWh a year ÷ 8.76 = 57 W.
  2. Add the station's own use. In CNET's test it ran from 12 W to 74 W. If you don't know a model's figure, the middle of CNET's results was about 31 W.
  3. Divide usable capacity by the total. The Anker SOLIX C2000 Gen 2 has 1,812 Wh usable and uses about 12 W itself: 1,812 ÷ (57 + 12) = about 26.2 hours. On CNET's fridge, which used a little less, it ran for 27.40 hours.

Real use often runs higher than the label, because door openings and a warm room add to it.

Work it out in our run-time calculator: pick your power station, add your fridge's kWh a year, and add solar panels if you have them.

Can a portable power station handle a fridge's start-up surge?

Yes, at these sizes. A compressor draws a short burst of extra power each time it starts. All 19 units in CNET's test started and ran its fridge, which is rated at 7.2 A at 118 V (about 850 W at full load).

Do I need a UPS mode to run a fridge?

Not to run it, but it helps if the fridge stays plugged into the station. In UPS (pass-through) mode the fridge runs from the wall, and the station takes over when the power goes out. Nobody needs to be home to switch it.

How long does food stay safe without power?

Not long without a backup. FoodSafety.gov gives these limits:

Every unit in CNET's test ran the fridge for at least 12.58 hours, more than 3 times the 4-hour limit.

Which portable power station is best for a refrigerator?

It depends on how long you need to cover and what you want to spend. These picks come straight from the table.

We may earn a commission from the store links below; it never changes our figures. How we make money

Skip for a fridge: the Oscal PowerMax 2400. It ran the fridge for 12.58 hours, the shortest of any model we track.

Can I run a mini fridge or a 12 V car fridge instead?

We don't have tests for those yet. A 12 V compressor fridge can run from a station's 12 V car outlet. That skips the inverter, which is where much of the overhead above comes from.

Sources and how we worked it out

We record CNET's numbers only. The testing, the write-up and any photos belong to CNET. How we test and rank