Which power station size fits your load?

A power station has two ratings that sound similar and behave differently. The Wh figure is the size of the tank. The watt figure is the width of the pipe. A unit with a large tank and a narrow pipe runs a small load for a long time and cannot start a large one at all.

The surge rating is the third number, and it is the one most often missing from the listing. A compressor or a pump needs several times its running current for a fraction of a second, and an inverter that cannot deliver that spike will simply shut down. The appliance then appears broken when the station is the thing that refused.

Sizing therefore runs three checks in order. Capacity for the hours you need. Continuous output for the load plus a margin. Surge for the biggest motor in the group. A station that fails any one of the three is not a smaller version of the right answer; it is a different product.

The calculator in the right-hand panel runs all three and names the smallest unit in our list that clears capacity and output, which is usually a useful surprise.

The most expensive mistake in this category is sizing to the house. A house with a 200 A service and central air draws far more than anyone backs up, and buying a station that covers it means buying a fixed system instead. The load that matters is the one you would actually keep running.

In practice that list is short. A fridge cycling on and off, a few LED bulbs, a router, a phone charger and maybe a fan come to roughly 300 W of average draw. That figure is the one to size against, and it is why a 2 to 3 kWh station covers a real outage for most households.

That list totals 2,340 Wh a day with a peak of about 315 W and a fridge startup of roughly 600 W. It is a real outage kit, not a hypothetical one, and it fits inside a single 3 kWh station with margin.

It is worth writing your own version of that table before looking at any product, because the act of writing it usually removes one appliance. A space heater at 1,500 W adds more to the requirement than everything else on the list combined, and most people decide against it once they see the number written down rather than imagined.

Two entries are commonly overestimated. A fridge is rated by its compressor, not by its plate, so the 150 W figure already accounts for the duty cycle when multiplied by eight hours. A router draws almost nothing but runs constantly, which is why it contributes 360 Wh despite being the smallest number in the running column.

Capacity is the running load multiplied by the hours you want it to run, then divided by the depth of discharge you are willing to use. At 85 %, a 300 W load for eight hours needs 2,824 Wh of nameplate, not 2,400. The 15 % buffer is not waste; it is what keeps the pack from being run flat every outage.

Read the last row and the category boundary appears: at twelve hours, the nameplate needed is 4,235 Wh, and the largest single unit we track is 4,096 Wh. That is not a gap you can close by shopping harder; it is the point where the answer changes shape.

The same table run backwards is the useful diagnostic. Divide a station's capacity by your running load and you get hours, then divide by your hours per day and you get days. A 2,048 Wh unit on a 300 W kit is about 5.8 hours of run at 85 % depth of discharge, which is most of one night but not a full day of the kit described above.

Continuous output is compared with the running load plus a margin. A 300 W load wants about 375 W of continuous rating, which any station above the smallest tier provides. This limit is rarely the binding one for a house kit.

Surge is where the surprises are. A fridge with a 150 W compressor needs roughly 600 W to start, so the surge requirement is about four times running watts, not two. A pump is worse. The same 300 W kit with a well pump added can need 2,000 W of surge or more, and that single appliance then decides the purchase.

Published surge figures vary in meaning. Some brands quote a peak for a fraction of a second, others quote a motor-start figure for several seconds, and many quote nothing. Treat an unpublished surge rating as equal to the continuous rating, then decide whether that is enough for the motors you own.

If the motor is the whole reason for the purchase, the fridge runtime calculator works the same startup problem for the most common appliance in the kit.

Appliance labels are nominal figures measured under test conditions. A motor draws more as bearings wear and as voltage sags. An inverter asked to deliver 100 % of its rating for hours runs hot and loses efficiency, so the useful output is less than the badge. A heater rated 1,500 W may draw a little more than that at the low end of the voltage tolerance range.

A 1.25 margin absorbs all of that without moving you up a whole product tier in most cases. It is the same multiplier used elsewhere on this site for continuous current, and it exists for the same reason: the label describes a fresh appliance on a good day.

Where the margin does move a tier, it is usually doing you a favor. A kit that sits exactly at the rating has no room for the second appliance you will add during the first real outage.

A worked case makes the margin concrete. A 1,200 W kettle run for six minutes is 120 Wh of energy, but it is 1,500 W of demand once the 1.25 margin is applied, so a station rated exactly 1,500 W continuous is the floor rather than a comfortable choice. Leave a 60 W router and a 40 W lamp running at the same time and the load becomes 1,300 W, which with the margin asks for 1,625 W and pushes you into the next tier. Appliances that never run together do not have to be added up: the number that matters is the peak, not the sum of every rating printed on a label.

Past about 4 kWh of nameplate, the answer stops being a bigger portable station. Four options remain, and they are not equivalent.

The last row is not a joke. Dropping a space heater or a hair dryer from the backup plan often brings a 5 kWh requirement back under 4 kWh, which is the difference between a portable purchase and a construction project.

Parallel operation deserves a warning because it is advertised loosely. Two units in parallel usually means the manufacturer sells a cable and supports the feature on specific models; it is not a property of the category. Check the model, not the brand, and check whether the pair adds capacity, output, or both.

Fourteen units in the current US market, ordered by capacity, with the continuous output beside it. Match your required capacity and your required output against both columns; the surge column matters only for the motors in your kit. For the wider comparison, see the portable power station ranking.

If the load you are sizing is a solar array rather than a battery, the battery sizing calculator starts from generation, and the unit converter handles the amp-hour confusion that comes with it.

Sizing to the house instead of the kit. The single most expensive error in the category, and the easiest to avoid by writing the short list first.

Buying on watt hours only. Capacity without output is a battery that cannot start your biggest appliance. Check both columns.

Assuming surge is twice the output. For a fridge it is closer to four times running watts. For a pump it can be worse.

Forgetting the depth of discharge. Planning on the nameplate number makes every runtime about 15 % optimistic before losses.

Planning to expand later with a different brand. Expansion batteries are brand-locked and often line-locked. If growth is likely, buy the ecosystem rather than the single unit.

Every figure here can be recomputed: energy is load times hours, nameplate is energy divided by depth of discharge, output is load times the margin. If a number disagrees with the calculator, the calculator is the source of truth, and the changelog records each change.

Three numbers are enough to reject most wrong listings before reading another word. The capacity in Wh, the continuous output in W, and the surge rating if it is published at all. If any of the three is missing, the listing is asking you to guess.

Start with your own kit total. Add up the running watts of everything you would keep on, multiply the biggest motor by four for its start, and decide how many hours you need. Those three figures are your requirement, and they do not change with the product.

Then compare in the same order the calculator uses. Capacity has to clear the hours; output has to clear the running total with margin; surge has to clear the biggest motor. A listing that clears the first and fails the second is a battery with a small inverter, which is the most common mismatch sold as a bundle.

Two shortcuts save time. Capacity divided by your running total gives hours, so a 1,024 Wh unit on a 300 W kit is about 2.9 hours of usable run, not eight. Continuous output divided by the biggest single appliance gives the headroom for that one appliance, which is the number that decides whether a kettle is allowed in the plan.

And check the units themselves. A listing that mixes Wh with Ah without giving a voltage is telling you nothing about run time, which is the confusion the unit converter exists to clear up.

If the outage lasts days rather than hours, the recharge balance matters more than the capacity: see solar generators for home backup.

Deciding between a battery box and a grid-tied kit at all? See plug-in solar versus a power station.

Comparing a battery with a fuel generator? See solar generator versus gas generator.