Power stations from 1,500 to 2,500 W, compared on published specifications

Enough to run a microwave, a circular saw, or a coffee machine — the point at which a station handles the things that briefly draw a great deal.

We do not test power stations. The median high-output station here holds 1,843 Wh and runs 46 min at its own rated output — the figures worth carrying into any listing you read.

Every figure below is taken from the manufacturer’s published listing and measured against the rest of the class. A watt-hour capacity is published by 91% of the high-output power stations here; where a listing is silent, the page says not stated rather than filling the gap.

Where the data runs out. 2 capacity bands hold too few listings for a median: 2,500 – 5,000 Wh (4), 5,000 Wh and up (5). Below 8 a median describes the sample rather than the market.

What the numbers say about this class

Baselines for this output class. A specification needs at least 8 listings stating it before a median is published.
Specification	Stated	Median	Middle half	Full range
Capacity	42	1,843 Wh	1,024 Wh – 2,048 Wh	768 Wh – 12,000 Wh
Continuous output	46	2,000 W	1,800 W – 2,400 W	1,500 W – 2,400 W
Runtime at full output (derived)	42	46 min	34 min – 1h 15m	23 min – 5h 27m
Energy per kilogram (derived)	33	91 Wh/kg	85 Wh/kg – 114 Wh/kg	73 Wh/kg – 652 Wh/kg
Weight	37	28 lb	26 lb – 41 lb	22 lb – 58 lb
Solar input	24	400 W	200 W – 600 W	40 W – 2,400 W

The middle-half column is the more useful one. It spans the 25th to 75th percentile, so it describes where most of the class actually sits rather than being pulled about by one unusual unit.

Watts and watt-hours are not the same question

Watts decide whether an appliance runs. Watt-hours decide how long. The box prints both and rarely explains that the first is a hard ceiling: a station rated below an appliance’s draw will not start it for one second, however large its battery.

Median capacity, median output, and the runtime those two imply — by output class.
Output class	Listings	Median capacity	Median output	Runtime at full output
Essentials Power Stations	64	231 Wh	300 W	52 min
Appliance Power Stations	25	999 Wh	1,000 W	1h 0m
High-Output Power Stations	46	1,843 Wh	2,000 W	46 min
Home Backup Power Stations	33	3,600 Wh	3,600 W	51 min

Read across that table and the pattern is the point. Outside the largest capacities, the median station runs about the same time at full output whatever class it is in — because capacity and output rise together. A bigger station does not buy you a longer run at full tilt; it buys you the ability to run more, or bigger, for about the same hour.

The runtime figure is arithmetic, and it is optimistic. It is the published capacity divided by the published output, which is the best you could ever get. Real inverters are 85 to 90% efficient, so the usable figure is lower — and no listing on this site states its inverter efficiency, so we do not estimate it. Treat the number as a ceiling and a way to compare like with like, not as a promise.

How this class spreads across capacities

Where this class’s 46 listings sit by capacity.
Capacity band	Listings	Median capacity	Runtime at full output
under 500 Wh	0	None in this class
500 – 1,200 Wh	19	1,024 Wh	34 min
1,200 – 2,500 Wh	14	2,000 Wh	51 min
2,500 – 5,000 Wh	4	Too few for a median
5,000 Wh and up	5	Too few for a median
Not stated	4	The listing publishes no watt-hour figure

How to choose one

Heating elements are the honest test

A microwave, a kettle, a hair dryer and a space heater all convert electricity straight to heat, which means they draw their full rating continuously with no cycling and no mercy. A 1,800 W kettle on a 2,000 W station is running that station at 90% of everything it has, and a 2,000 Wh battery will empty in about an hour. This class makes such appliances possible; it does not make them cheap in watt-hours.

Watch the socket the station charges from

A station that outputs 2,000 W frequently accepts far less coming in. Recharging from a domestic socket is limited by that socket long before it is limited by the battery, and a large station on a slow input can take most of a day to refill. Where a listing states an AC input rating it is worth more attention than the output figure most buyers read first.

Weight has stopped being incidental

At this size a station is typically 20 to 30 kilograms, which is luggable rather than portable. If it will be moved regularly rather than sited once, the energy-per-kilogram figure and the presence of wheels or handles matter more than the last few hundred watts of output.

Where the weight becomes the specification

At this output a station stops being something carried casually and becomes something moved deliberately. Weight scales almost linearly with capacity across this corpus, and energy density varies far less between models than the marketing suggests, so a heavier station in this class is generally a bigger one rather than a worse-built one. The practical questions are whether it will be lifted into a vehicle, whether it has wheels and a handle, and whether the place it lives is up any stairs — none of which appear in a specification table, and all of which decide whether it gets used.

Fans, and where the noise turns up

A large inverter under load produces heat, and the fans that remove it are audible. Most stations are near-silent while idle or trickling and become noticeable when asked for serious power or when charging quickly from mains. This is rarely stated and never quantified, and it is the specification that decides whether a station can sit in a bedroom during an outage or has to live in a hallway. The general shape is that fan noise tracks how hard the unit is working rather than its size.

What the listings do and do not tell you

Specification coverage is uneven, and that unevenness is worth knowing before you shop. A missing figure is not a worse product — it is a less completely described one.

How many listings in this class publish each specification.
Specification	Stated	Share
Continuous output	46	100%
Capacity in watt-hours	42	91%
Surge rating	20	43%
Cell chemistry	39	85%
Solar input	24	52%
Weight	37	80%

Where a listing is silent, this site prints not stated. It never fills the gap with a plausible default.

Mistakes this data makes visible

Buying watt-hours to solve a watts problem

The most common and the most expensive. If a station cannot start your appliance, a larger battery in the same output class will not change that. Check the continuous rating against the appliance’s draw first, and the surge rating against its startup.

Reading a surge rating as a running rating

Surge is a brief peak while a motor starts, not a level anything can be run at. A fridge that runs at 150 W may demand 900 W for a fraction of a second, and a station whose surge headroom is thin will trip rather than start it.

Assuming solar means independence

A station accepting 200 W of panel gathers at best around a kilowatt-hour across a good day, and panels are almost always sold separately. Solar input is a ceiling on how fast it refills, not a promise that it will.

Find one that fits

The power station finder filters on capacity, output, chemistry and solar input, and shows each result against the median for its own capacity band — with its runtime at full output on every card.

Narrower comparisons

Same data, scoped to the group people most often shop for:

LiFePO4 High-Output Power Stations

Individual stations in this class

One page each, with the runtime its own figures imply and how it sits against the median above: