Battery backup for Starlink: what the dish actually draws
Battery backup for Starlink: Starlink publishes an average power figure for every kit. The sizing arithmetic, and the DC path that skips the inverter.
Amazon links below are affiliate links: if you buy through them we earn a commission at no extra cost to you. As an Amazon Associate, The Home Battery Report earns from qualifying purchases. Commissions never change a pick; rankings are set from the datasheets first.
2,764.8 Wh with 3,840 W maximum output and a published 6,000+ cycles to 80 percent. Against the 75 to 100 W average Starlink publishes for the Standard 4 X, that is a ceiling of roughly 36.9 hours at the low end of the range and 27.6 hours at the high end, before conversion losses. The smallest capacity in this group.
3,840 Wh with 6,000 W maximum output, delivered through its 240V-capable NEMA L14-30R port. Against the same 75 to 100 W figure that is a ceiling of roughly 51.2 hours at the low end and 38.4 hours at the high end, with far more output than a dish will ever ask for.
4,096 Wh rated 4,000 W continuous with an 8,000 W surge, expandable on one unit. Against Starlink's 75 to 100 W average that is a ceiling of roughly 54.6 hours at the low end of the range and 41.0 hours at the high end, with output to spare for loads far heavier than a dish.
6,144 Wh per battery with 7,200 W continuous, expandable to 30 kWh on one inverter. Against the 110 to 150 W Starlink publishes for the Performance kits, that is a ceiling of roughly 55.9 hours at the low end and 41.0 hours at the high end, the only unit here that clears two full days on any part of that range.
Every figure is a manufacturer datasheet spec. We have not lab-tested these units.
Most of the loads on this site have to be reverse engineered from a nameplate, because the manufacturer does not publish what the thing actually draws. Starlink is the exception. SpaceX publishes an average power figure for every kit generation it sells, in watts, on its own help center, and it tells you what the figure includes. That makes satellite internet one of the few backup sizing problems where both sides of the division are a published number. We have not tested any of these products, and the Amazon links above are affiliate links, disclosed at the top of this page.
What Starlink publishes
The help center article "How much power does my Starlink need?" lists four sets of figures. Read August 19, 2026 at starlink.com/support:
| Kit | Average | Idle |
|---|---|---|
| Starlink Mini | 20 to 40 W | 15 W |
| Standard Actuated | 50 to 75 W | 20 W |
| Standard 4, Standard 4 X, Enterprise | 75 to 100 W | 20 W |
| Performance (Gen 1) and Performance (Gen 2) | 110 to 150 W | 45 W |
Two qualifications come from Starlink itself, and both change how you use the table. The first: "The Starlink specs below include the Starlink, WiFi router, power supply, and cables." That is unusually generous. It means the number is a whole-kit figure at the wall, not a dish-only figure you then have to add a router to. The second: "Note that the specs are based on AC input power averages." So the figure already contains the power supply's own conversion loss, which is the loss you would otherwise have to guess at.
The Standard 4 X product specification sheet published at starlink.com/public-files agrees with the help center on the dish, listing "Power Consumption Average: 75 - 100 W" and rating the power supply at 100 to 240V, 2.5A, 50 to 60 Hz. The Performance kit specification sheet lists the same "Average: 75-100W" for its dish, which is lower than the 110 to 150 W the help center gives for the Performance kit as a whole. Where a maker publishes two figures, we print both and size against the larger one.
Turning that into daily energy
Multiply the average by 24 and you get the number that matters for a multi-day outage:
- Starlink Mini: 0.48 to 0.96 kWh per day.
- Standard Actuated: 1.2 to 1.8 kWh per day.
- Standard 4 X: 1.8 to 2.4 kWh per day.
- Performance: 2.64 to 3.6 kWh per day.
That top figure deserves a moment. GE publishes 700 kWh per year for its GFE26JYMFS French-door refrigerator, which works out to about 1.92 kWh per day. A Standard 4 X at the high end of Starlink's own published range uses more energy in a day than that refrigerator does, and a Performance kit uses nearly twice as much. If your outage plan was built around the fridge, adding satellite internet roughly doubles it. Our refrigerator sizing guide does that appliance arithmetic in full.
Runtime against the units we track
Rated capacity divided by published average draw. The assumptions are stated and they matter: these use each unit's rated Wh rather than usable Wh, and they ignore inverter losses, the low-voltage cutoff and the station's own self-consumption. Real runtime is lower.
| Unit | Rated capacity | Mini at 40 W | Actuated at 75 W | Standard 4 X at 100 W | Performance at 150 W |
|---|---|---|---|---|---|
| BLUETTI Apex 300 | 2,764.8 Wh | 69.1 h | 36.9 h | 27.6 h | 18.4 h |
| Anker SOLIX F3800 Plus | 3,840 Wh | 96.0 h | 51.2 h | 38.4 h | 25.6 h |
| EcoFlow DELTA Pro 3 | 4,096 Wh | 102.4 h | 54.6 h | 41.0 h | 27.3 h |
| Jackery Explorer 5000 Plus | 5,040 Wh | 126.0 h | 67.2 h | 50.4 h | 33.6 h |
| EcoFlow DELTA Pro Ultra | 6,144 Wh | 153.6 h | 81.9 h | 61.4 h | 41.0 h |
Read the shape rather than the decimals. A Mini is a load that any of these units carries for most of a week. A Standard 4 X is a load where the difference between a 2.8 kWh box and a 6 kWh box is the difference between one long day and two and a half. A Performance kit is heavy enough that only the largest units here clear two full days, which is the case for recharging rather than for buying more battery.
None of these makers publishes an inverter efficiency figure at a 100 W load, which is exactly the low-load region where inverters are least efficient. We do not apply a derate rather than invent one, but the honest read is that a small steady load like a dish will underperform the table by more than a large one would.
The DC path, and why it is worth asking about
The table above assumes the dish is plugged into an AC outlet on the station, which means the battery's DC is inverted to AC and then rectified back to DC inside Starlink's power supply. Starlink sells hardware that skips that.
Its support article on DC-input power supplies states that the Standard DC-DC Power Supply, available only in select markets, "allows you to power Starlink Standard directly from a DC power source, removing the need for an inverter when powered from your RV, car, boat, or battery pack," and that it uses XT60 connectors on both sides with 12 AWG wire. The Advanced Power Supply included with the Performance kit is described in that kit's specification sheet as rack mountable, operating "on both AC and DC power," and supporting "DC input with a backup battery for uninterrupted operation," rated 100 to 240 VAC, 12 to 56 VDC.
The Mini is the simplest case of all. Starlink's support article on supported power sources states plainly: "The Starlink Mini has a power input rating of 12-48VDC, 60W." That is a battery-friendly spec, and it is why a Mini is the kit people actually run off a small pack for days.
Starlink attaches a caveat to all of this that we will repeat rather than soften: "Starlink only guarantees performance with the included Starlink power supply and cable." A DC path removes a conversion penalty that nobody publishes a number for, and it also removes you from the configuration the manufacturer supports. That is a trade, not a free win.
When a smaller unit is the right answer
The five units above are sized for a house, not for a dish. If satellite internet is the only thing you are keeping alive, the same division points much smaller. A Standard 4 X at Starlink's 100 W figure consumes 1 kWh in ten hours, which means a one kilowatt hour class power station covers a working day and a two kilowatt hour class unit covers most of two. Those smaller classes are compared with their own published capacity and switchover figures in our work-from-home outage kit, which is the right page if the requirement is a desk rather than a household.
The case for the larger units on this page is that the dish is rarely the whole list. Once you add a refrigerator, some lighting and a phone charger, the dish stops being the load and becomes the reason the total is uncomfortable. That is the point where the plug-in backup comparison and the portable versus installed comparison are the useful next reads.
What to check before you buy
Find out which kit you own before you size anything, because the spread between a Mini and a Performance kit is more than seven times at the idle figure and nearly four times on the average. Starlink publishes a "What Starlink Kit do I have?" article for exactly this. Then decide whether you are sizing for a summer outage or a winter one: the Standard 4 X sheet lists a snow melt capability of up to 40 mm per hour and the Performance sheet up to 85 mm per hour, and Starlink says outright that consumption varies with temperature. Plan against the top of the published range if your outages come with weather.
Every figure above is a manufacturer claim recorded with its published basis: Starlink's help center power article and its Standard 4 X, Mini and Performance specification sheets, read August 19, 2026, and the power station capacities from each maker's own spec table. The runtime numbers are arithmetic on those claims under the assumptions stated, not measurements we made, and not a promise about your dish.