Starlink published the specification sheet for its V5 residential dish on 14 July 2026, and one number travelled faster than the rest of the document: average power consumption of 35 to 50 watts, against 75 to 100 watts for the Standard V4 it replaces.
The number is real. It comes straight from Starlink’s own specification sheet, and it represents the single largest efficiency improvement in the history of the consumer terminal. Anyone who has spent a winter watching a dish eat 2.5 kWh a day out of a battery bank should be pleased.
But in the three weeks since, that figure has been reproduced across dozens of off-grid and RV publications with almost no examination of what sits behind it. And there are three things behind it that change the conclusion — one of which is not a caveat at all, but an omission on the spec sheet itself.
What Starlink actually published
Start with the primary source, because a surprising amount of secondary coverage has garbled it.
| Starlink Mini | Starlink V5 | Standard (V4) | |
|---|---|---|---|
| Average power consumption | 25–40 W | 35–50 W | 75–100 W |
| Published DC input rating | 12–48 V, 60 W | none published | none published |
| First-party DC supply available | native (barrel jack) | no | yes, sold separately |
| Snow melt capability | up to 25 mm/hr | up to 40 mm/hr | up to 40 mm/hr |
| Wind speed (operational) | 96 kph+ (60 mph+) | 265 kph (165 mph), mounted | 96 kph+ (60 mph+) |
| Dish weight | 1.10 kg (2.43 lb) | 1.1 kg (2.4 lb) | 2.9 kg (6.4 lb) |
| Dish dimensions | 298.5 × 259 × 38.5 mm | 384 × 306 × 34 mm | 594 × 383 × 39.7 mm |
| Rated peak download | not published | 375+ Mbps | 400+ Mbps |
| Bundled router | integrated, Wi-Fi 5, 3×3 | Router Mini, Wi-Fi 6, 2×2 | Gen 3, Wi-Fi 6 |
| Rated for in-motion use | yes | no | yes (Standard 4 / 4X) |
Figures from Starlink’s published specification sheets for the V5, Mini and Standard kits, retrieved August 2026.
Two things in that table deserve more attention than the wattage. The V5’s mounted wind rating of 265 kph is nearly three times the Standard’s published figure — a genuinely significant improvement for exposed rural sites that nobody wrote a headline about. And the snow melt capability is identical to the V4’s.
Hold on to that second one.
Finding one: it is an average, and there is no published peak
Starlink lists a single line for power: an average range. It does not publish an idle figure, a boot-surge figure, or a snow-melt figure for the V5. It never has for any terminal.
This matters because off-grid systems do not fail on averages. They fail on moments — an inverter tripping on a surge, a battery hitting low-voltage cutoff in a cold snap, a controller sized to a daily total that never anticipated a sustained heater load. None of those are visible in a 35-to-50-watt range.
The heater is the specific problem. Every Starlink terminal since the second generation carries a resistive element behind the phased array that activates automatically when the dish detects moisture at or near freezing. Starlink does not publish its draw. Independent measurements from cold-climate users vary widely — some report an additional 20 to 30 watts sustained, others report excursions well past 100 watts during active storms. The spread is wide enough that anyone quoting a single number is guessing.
What we can say with confidence is structural: the V5’s snow melt capability is rated to the same 40 mm per hour as the V4. Melting a given quantity of snow takes a given quantity of energy, and the phased array behind it did not get smaller in area by nearly as much as the electronics got more efficient. There is no basis for assuming the winter penalty halved along with the average draw.
For a northern install, this means the ratio between your summer and winter consumption gets worse on a V5, not better. The baseline dropped; the heater load did not.
Finding two: there is no published DC input rating
This is the omission, and it is the one that matters most to this audience.
Look again at the table. The Mini’s specification sheet publishes an input rating: 12 to 48 volts, 60 watts. That single line is why the Mini became the default dish for vehicles and off-grid cabins — you can wire it to a battery bus and skip AC conversion entirely.
The V5 specification sheet publishes no input voltage at all. It documents the power supply’s dimensions, its 0.40 kg weight, its IP66 rating and its operating temperature range. It does not state what the brick accepts or what it delivers to the dish.
The V5 also moves to an external power supply architecture — a separate weather-rated brick, rather than power passing through the router as on the V4 kit. The Starlink cable from the dish and the Ethernet cable to the Router Mini both terminate at that brick, with mains feeding it. Out of the box, the V5 is an AC device.
This dish family has historically run on roughly 56 volts over the Starlink cable, and Starlink sells a first-party DC-DC supply for the Standard that accepts 12 to 48 volts and outputs 56 volts at 3.57 amps. Third-party vendors who build DC conversion hardware for the Standard say they expect their units to work with the V5, reasoning that the cabling and power architecture carried over — while acknowledging that Starlink has not published the V5 brick’s electrical specifications and that they are still validating against production hardware.
That is an honest position. It is not a specification. As of early August 2026 there is no first-party DC power path for the V5 and no vendor-confirmed third-party one either. If you are planning a battery-native install around this dish today, you are planning around an unverified assumption.
Finding three: halving the load makes conversion overhead proportionally worse
Here is the part that no coverage has worked through, and it is arithmetic rather than opinion.
If you cannot feed a dish DC, you run it through an inverter. Inverters do not have a single efficiency figure — they have an efficiency curve, and that curve collapses at low loads because a fixed quantity of overhead is being spread across a smaller output. A pure sine inverter sized for a vehicle or cabin might hit 90% or better at several hundred watts. At 50 watts of output, 75 to 85% is a more realistic planning range.
So the fixed cost of conversion is roughly constant in watts, but the load it is attached to just halved. The percentage penalty therefore roughly doubled.
Working it through with planning assumptions of 80% inverter efficiency at V5-level loads, 85% at V4-level loads, and 92% for a DC-DC path, and adding a few watts for the router:
| Setup | Draw at the battery | Energy per day |
|---|---|---|
| Mini, native DC | ~35 W | ~850 Wh |
| V5, via inverter | ~59 W | ~1,425 Wh |
| V5, via DC (if confirmed) | ~52 W | ~1,240 Wh |
| V4, via inverter | ~109 W | ~2,610 Wh |
| V4, via first-party DC-DC | ~101 W | ~2,415 Wh |
Planning figures using the midpoint of each published average range, continuous operation, clear conditions, no snow melt. Your measured draw will differ.
Two conclusions fall out of this.
The first is that the V5’s advantage is genuine but smaller than the spec sheet ratio implies. The published averages sit in a 2.06:1 ratio. Run both through inverters and you land nearer 1.83:1 — a 45% reduction in daily energy, not the 50%+ the headline numbers suggest. Still an excellent result. Just not the one being advertised.
The second is that the DC-versus-AC decision now matters more than it did, not less. Moving a V4 from an inverter to Starlink’s DC-DC supply saves around 8% of its consumption. Moving a V5 to a DC path, if one is ever confirmed, would save around 13%. The lighter the load, the larger the share the inverter takes.
Finding four: the people most excited about this dish cannot use it
The V5 is sold on Residential service. Starlink’s documentation is direct about the boundary: in-motion use is not permitted on Residential plans, and the V5 does not appear on the list of hardware supported for in-motion operation — which currently covers the Performance terminals, the Standard 4 and 4X, and the Mini.
This has been badly reported. A dramatically lighter, more efficient dish reads like an obvious win for vans, trailers and boats, and a good deal of coverage has framed it that way. It is not. The V5 is fixed-site residential hardware with no in-motion rating, no published DC input, and a plan restriction on top. For anything that moves, the Mini on a Roam plan remains the answer.
Where the V5 does land well is the fixed rural property: a cabin, a homestead, an outbuilding, a permanently sited trailer. There, mains or a large bank is usually available, an inverter is already running for other loads, and the wind rating and reduced roof load are real advantages alongside the power saving.
Who is still publishing wrong numbers
Two categories, and they fail differently.
The first is guides written before July 2026 and never touched since. An article telling you a standard dish draws 75 to 100 watts is describing hardware that is still sold and still current for most plans — not wrong, just no longer complete. Search results do not make that distinction, so a reader landing on a 2025 sizing guide will size an array for a dish that has since been superseded.
The second is more recent and more misleading: coverage that took the 35–50W figure, applied it directly to a battery calculation, and published runtime estimates. Those numbers omit conversion losses entirely, and as shown above, conversion losses are proportionally larger on this dish than on any residential terminal that preceded it. Runtime figures derived that way run optimistic by roughly a quarter.
There is also a rollout problem most coverage glosses. The V5 opened in a handful of rural US areas in July — Idaho, Montana and Oregon among the first — tied to the entry-level Residential plan, with wider availability promised as production scales. Sizing a system around a dish you cannot yet order is a planning exercise, not a purchase decision.
What to do with this
If you are running a Starlink terminal off-grid today, the practical implications are narrow and specific.
Do not resize an existing system around the V5 until you can order one. Availability is regional and still expanding. A system sized for your current dish will run a V5 comfortably; the reverse is not true.
Measure, do not assume. An inline DC watt meter on the battery side of your inverter costs very little and settles every question in this article for your install, your climate and your usage.
Budget for winter separately. Whatever the heater costs you on your current dish is roughly what it will cost you on a V5. Size the winter case explicitly rather than applying a percentage uplift to an annual average. Our guide to sizing solar and battery for continuous Starlink uptime works through that stress case, and the wider question of whether to add panels or batteries first usually resolves toward the array in exactly this scenario.
Treat the DC question as open. If a battery-native V5 install is central to your plan, wait for confirmed hardware. It may well arrive. It does not exist yet.
Do not buy hardware to save electricity. A V5 saves roughly 1.2 kWh a day over a V4. On mains, at an average US rate, that is a few dollars a month and will not repay a kit purchase in any sensible horizon. Off-grid, where the saving is denominated in battery capacity and panel area rather than cents, the calculation is entirely different and often does favour the upgrade. That distinction applies to most connectivity spending: our five-year comparison of off-grid internet costs treats hardware, service and power as one number, which is the only way it makes sense. For anyone running a rig rather than a fixed site, connectivity is one line in a larger picture — the itemised five-year cost of RV ownership puts it in proportion.
If you are choosing between terminals rather than powering one you already own, the comparison is more involved than a wattage figure, and we have worked through which dish suits which setup in detail.
Frequently asked questions
How many watts does the Starlink V5 use? Starlink publishes an average of 35 to 50 watts. That figure excludes snow melt, does not include a published peak or idle value, and appears under the dish’s specifications rather than as a whole-system figure. Budget a few additional watts for the Router Mini, whose consumption Starlink does not publish.
Is the V5 half the power of the V4? On published averages, close to it — the ranges sit at roughly 2:1. In a real off-grid install the gap narrows to around 45%, because inverter losses take a proportionally larger bite out of a smaller load.
Can I run a Starlink V5 on 12V or 24V DC? Not with any officially documented method. Starlink publishes no input voltage rating for the V5 and does not currently sell a first-party DC supply for it. Third-party vendors expect their Standard-compatible converters to work on the reasoning that the power architecture carried over, but as of August 2026 that is an expectation rather than a confirmed specification.
Does the V5 use less power in winter? Its baseline is lower, but its snow melt capability is rated identically to the V4 at up to 40 mm per hour. There is no published basis for expecting a proportionally lower heater load, and cold-climate users should size for a winter penalty broadly similar to the V4’s in absolute watts.
Can I use a V5 in an RV or van? No. Starlink does not rate the V5 for in-motion use and does not permit in-motion operation on Residential plans, which are the only plans the V5 currently ships on. The Mini remains the correct choice for anything mobile.
Should I upgrade from a V4 to a V5 for the power saving? Off-grid, quite possibly — the saving is around 1.2 kWh a day, which translates into meaningful battery and array capacity. On mains, almost certainly not; the electricity saving is a few dollars a month.