Reference
Device Wattage Library
Source-labeled planning ranges for 23 devices people actually back up — with the evidence type, limitations, and official links kept beside the numbers.
Method and freshness: version 1.0.0, reviewed August 4, 2026. Most rows are source-informed editorial planning envelopes; a manufacturer rating, adapter ceiling, or named-model example is never relabeled as a PowerPreflight measurement. Verify your own label or measure the device before buying.
Every figure below is a planning range, not a measurement of your device. Each row says whether its endpoints are editorial or manufacturer-published. That is useful for planning — and the “Estimate runtime” link on each row drops the typical figure straight into the Battery Runtime Calculator. But before you spend real money on a battery, it’s worth pinning down your own device’s real draw.
Backing up more than one device? Build your full load in the Multi-Device Load Builder — it totals watts and daily watt-hours across everything you pick (fridge duty cycles applied automatically) and turns the sum into a station size and runtime estimates.
How to verify your device’s real wattage
- Check the device label or power brick. Most gear lists its power input near the plug. Note this is usually a maximum rating — real draw is often lower. If it only lists amps and volts, watts ≈ amps × volts.
- Look up the manufacturer’s spec sheet. Search the exact model number plus “specifications.” Spec sheets often list typical and maximum consumption separately, which is exactly the split you want.
- Measure with a plug-in watt meter. A basic meter sits between the device and the outlet and shows the actual draw as you use the device. It’s the only method that captures how your settings and usage change the number — and it settles the question for good.
Some links on this page may be paid links. If you buy through them, Cynosure LLC may earn a commission at no extra cost to you. We do not claim to have personally tested products unless clearly stated.
Our pick is the 20-year default: P3 P4400 Kill A Watt Electricity Usage Monitor ($25–$35)— price last checked 2026-07-09, not hands-on tested by us.
Our assumptions and formulas are documented on the How We Estimate page. When you know your watts, the Power Station Sizing Calculator turns them into a capacity recommendation.
No devices match that search. Try a shorter word, like “fridge” or “laptop”.
Connectivity & internet
Modems, routers, hotspots, and satellite terminals — the loads most people back up first, and usually the cheapest to cover.
| Device | Typical watts | Range | Notes | Runtime / plan |
|---|---|---|---|---|
| Wi-Fi router | 10W | 5–15W | Editorial class range. One cited legacy router uses a 9V × 0.6A adapter (5.4W output ceiling); newer and mesh devices vary. Source-informed editorial planning rangeLimits: A single adapter-output ceiling does not establish a measured 5–15W router-class range. Official evidence:
| Estimate runtimePlan internet backup |
| Modem | 8W | 5–15W | Editorial planning range for cable modems and fiber terminals. Back up both modem/terminal and router unless one gateway combines them. Source-informed editorial planning rangeLimits: The cited modem manual lists a 30W adapter ceiling, not measured consumption or this class range. Official evidence:
| Estimate runtimePlan internet backup |
| Router + modem combo | 15W | 10–25W | Editorial planning range for an ISP gateway. A replacement adapter can be rated far above the gateway's ordinary draw. Source-informed editorial planning rangeLimits: The cited 42W replacement adapter is shared by several products and is not a gateway-consumption measurement. Official evidence:
| Estimate runtimePlan internet backup |
| Mobile hotspot | 5W | 2–10W | Editorial planning draw while in use and charging. A cited mobile Wi-Fi example accepts 5V/1A input, a 5W ceiling rather than measured use. Source-informed editorial planning rangeLimits: One model's input ceiling does not establish the class endpoints or actual in-use charging draw. Official evidence:
| Estimate runtime |
| Satellite internet terminal | 75W | 50–100W | Planning range based on named Starlink Standard generations: official support publishes 50–75W and 75–100W AC-input averages including terminal, router, supply, and cables. Source-informed editorial planning rangeLimits: The cited ranges apply to named Starlink Standard generations, not every satellite internet terminal. Power varies with temperature, location, and use. Official evidence:
| Estimate runtimePlan satellite backup |
| Starlink Mini-style satellite internet terminal | 30W | 20–40W | Official Starlink support publishes a 20–40W average range for Mini; its separate specification sheet says 25–40W average and a 60W input rating. Manufacturer-published range for the named exampleLimits: The user-facing 20–40W range follows the support article; the 25–40W specification-sheet discrepancy is preserved rather than hidden. Official evidence:
| Estimate runtimePlan satellite backup |
Computing & work
Laptops and work gear. Draw swings a lot with workload, so the ranges here are wider than for most categories.
| Device | Typical watts | Range | Notes | Runtime / plan |
|---|---|---|---|---|
| Laptop | 50W | 30–100W | Planning envelope for active use plus charging. Workload and battery state can move real wall draw well below the power adapter's rating. Source-informed editorial planning rangeLimits: Apple's recommended adapter classes are capacity ratings, not workload consumption measurements. Official evidence:
| Estimate runtime |
| Gaming laptop | 180W | 100–360W | Conservative planning envelope for gaming load. Current Alienware examples ship with 130–360W adapters; the adapter rating is not the same as measured AC draw. Source-informed editorial planning rangeLimits: The official source documents required adapter classes, not actual wall consumption across gaming laptops. Official evidence:
| Estimate runtime |
| Portable monitor | 10W | 5–18W | Editorial range for USB-powered 14–16 inch panels. A cited 15.6-inch ASUS example publishes 7.6W consumption and a separate 15W input ceiling. Source-informed editorial planning rangeLimits: One model's published consumption does not establish the full 5–18W class range. Official evidence:
| Estimate runtime |
| External hard drive | 8W | 5–20W | Editorial envelope for portable and desktop external drives. A cited bare 3.5-inch drive publishes 6.7W idle and 10.5W maximum random-read consumption. Source-informed editorial planning rangeLimits: A bare-drive component figure does not include enclosure, USB bridge, or wall-adapter losses and does not establish the class endpoints. Official evidence:
| Estimate runtime |
Small electronics
Phones, chargers, lights, and entertainment devices. Individually tiny — but they stack, so count each one you plan to run.
| Device | Typical watts | Range | Notes | Runtime / plan |
|---|---|---|---|---|
| Smartphone | 10W | 5–20W | Planning draw while charging. Adapter ratings show available output, not the phone's measured wall draw; real draw tapers as the battery fills. Source-informed editorial planning rangeLimits: Apple publishes adapter classes, not a cross-brand phone-consumption range or energy per full charge. Official evidence:
| Estimate runtime |
| Tablet | 18W | 10–30W | Planning draw while charging. Larger tablets and fast-charge modes trend higher, but the adapter's rating is only an output ceiling. Source-informed editorial planning rangeLimits: The cited Apple adapter classes do not establish this full device-class range or a typical wall draw. Official evidence:
| Estimate runtime |
| LED light | 8W | 3–15W | Editorial range for one LED bulb or lantern. A cited Philips lamp publishes exactly 8W; count every separate light in a multi-light setup. Source-informed editorial planning rangeLimits: The named 8W lamp supports the exemplar, not every bulb, lantern, panel, or string-light endpoint. Official evidence:
| Estimate runtime |
| Camera battery charger | 15W | 5–30W | Planning range widened for single- and dual-bay chargers. Sony's BC-ZD1 guidance recommends 30W, works more slowly at 18W, and rejects 5V/1.5A input. Source-informed editorial planning rangeLimits: The official figures are input requirements for one dual-battery charger, not measured wall draw for the whole class. Official evidence:
| Estimate runtimePlan camera-camp power |
| Gaming handheld | 25W | 10–45W | Editorial whole-device charging-and-play envelope. Steam Deck publishes a 4–15W APU range and bundles a 45W supply; those are different quantities. Source-informed editorial planning rangeLimits: Component power and adapter capacity do not directly establish whole-device AC consumption. Official evidence:
| Estimate runtime |
| USB-C charger (common 18–100W class) | 45W | 18–100W | Scoped editorial range for common ≤100W phone/laptop adapters. The connected device negotiates the actual draw; higher-power 140W workstation adapters are outside this row. Source-informed editorial planning rangeLimits: Adapter wattage is available output, not measured wall consumption. Apple's cited lineup includes 140W adapters excluded by this row's explicit scope. Official evidence:
| Estimate runtime |
| Portable projector | 50W | 13–90W | Planning range widened to include official portable examples: BenQ publishes 13.1W eco/16.7W normal/24W maximum for GV1 and 85W typical for GV50. Source-informed editorial planning rangeLimits: Two named models establish a 13.1–85W exemplar span, not a complete portable-projector class range. Official evidence:
| Estimate runtime |
Comfort
Fans and similar comfort devices. Speed settings change the draw substantially, which matters over a long outage.
| Device | Typical watts | Range | Notes | Runtime / plan |
|---|---|---|---|---|
| Small fan | 30W | 15–50W | Editorial class range. Named Vornado examples publish 15/30W and 33/39/45/49W speed ratings; other USB, desk, and box fans vary. Source-informed editorial planning rangeLimits: Two manufacturer examples support a 15–49W exemplar span, not the endpoints of every small-fan type. Official evidence:
| Estimate runtime |
Appliances
Refrigeration, pumps, and TVs. A Surge badge means startup can exceed running watts; verify the exact multiplier and peak-output requirement for your model.
| Device | Typical watts | Range | Notes | Runtime / plan |
|---|---|---|---|---|
| Mini fridgeSurge | 60W | 40–100W | Editorial running-watts and duty-cycle assumptions. Compressor startup can exceed running draw; verify the data plate and measure a full cycle before buying. Source-informed editorial planning rangeLimits: ENERGY STAR defines compact-refrigerator categories and annual-energy testing, not running watts, duty cycle, or startup surge. Official evidence:
| Estimate runtimeBuild an outage load |
| Full-size refrigeratorSurge | 150W | 100–250W | Editorial running-watts and duty-cycle assumptions. Compressor startup can exceed running draw, so verify both continuous and surge requirements for your exact model. Source-informed editorial planning rangeLimits: ENERGY STAR's category/test-method page does not publish this running range, duty cycle, or a startup-surge value. Official evidence:
| Estimate runtimeBuild an outage load |
| Residential sump pump (generic sizing example)Surge | 900W | 800–1050W | Generic generator-sizing estimate, not pump laboratory data. Generac lists 800–1,050 running watts and advises allowing 3× listed watts for motor starting; verify the pump label. Manufacturer-published generic sizing estimateLimits: The source is a generic generator-sizing table, not a model-specific pump test, and cycle frequency still depends on water inflow. Generic surge-planning figure: 3,150W. Verify the exact device before relying on it. Official evidence:
| Estimate runtimePlan pump backup |
| Small TV | 40W | 30–60W | Editorial range for 24–32 inch LED sets. A cited Samsung 32-inch model publishes 38W typical, 50W maximum, and 0.5W standby. Source-informed editorial planning rangeLimits: One exact 32-inch model supports an exemplar inside the range, not the endpoints for every 24–32 inch TV. Official evidence:
| Estimate runtime |
Health devices
Medical equipment is the one category where an estimate is not good enough — verify exact requirements before relying on battery power.
| Device | Typical watts | Range | Notes | Runtime / plan |
|---|---|---|---|---|
| CPAP-style deviceMedical | 55W | 30–75W | Illustrative planning envelope only. An AirSense 11 guide publishes 56.1W typical and 73.2W peak under its documented test; settings and heated accessories change demand. Source-informed editorial planning rangeLimits: One model-level test does not establish a CPAP-class range or your prescribed setup's requirements. Official evidence:
| Estimate runtimePlan CPAP backup |
Wattage figures for medical devices are illustrative only. Always confirm power requirements with the device manufacturer and your doctor before relying on battery backup. This is not medical advice.
Evidence
Source register and correction log
Official sources document named examples, ratings, or measurement methods. They do not automatically prove a whole device class. That boundary is preserved in every row and in both downloadable files.
Measurement method
U.S. Department of Energy, Federal Energy Management Program: Measuring Standby Power — Stable loads may use direct watts after a five-minute stability check; fluctuating loads require accumulated watt-hours divided by time. Measurement method only; it publishes no numeric values for the 23 device classes.
Apple · Apple USB power adapter examples
About Apple USB power adapters
Published value: iPhone 5W legacy and 18W fast-charge compatibility; iPad 10W, 12W, 18W, and 20W adapter classes.
Evidence kind: adapter-output
Limits: Adapter output and compatibility, not measured phone/tablet wall draw or watt-hours per charge.
Open the official source · checked 2026-08-04
Apple · Mac notebook adapter recommendations
Use a power adapter with your Mac
Published value: Recommended classes include 20W, 30W, 35W, 40–60W, 67W, 70W, 87W, 96W, and 140W depending on Mac model.
Evidence kind: adapter-output
Limits: Recommended adapter capacity, not workload draw or AC wall-input consumption.
Open the official source · checked 2026-08-04
Dell · Current Alienware laptop examples
No Power after Shut Down Using the Incorrect AC Adapter on Alienware Gaming Laptops
Published value: Shipped adapter classes include 130W, 240W, 280W, 330W, and 360W.
Evidence kind: adapter-output
Limits: Required adapter size is not a direct measurement of gaming-load AC consumption.
Open the official source · checked 2026-08-04
TP-Link · TL-WR841N v1
TL-WR841N 300Mbps Wireless N Router
Published value: 9V × 0.6A adapter output, a 5.4W ceiling.
Evidence kind: volts-amps-ceiling
Limits: One legacy model and an adapter ceiling, not measured consumption or a router-class range.
Open the official source · checked 2026-08-04
NETGEAR · CM1000 cable modem
CM1000 User Manual
Published value: 12V × 2.5A adapter output, a 30W ceiling.
Evidence kind: volts-amps-ceiling
Limits: Adapter ceiling for one cable modem, not measured draw or a modem-class range.
Open the official source · checked 2026-08-04
NETGEAR · PA7000P replacement adapter, including C7000 compatibility
PA7000P Power Adapter
Published value: 12V × 3.5A output, a 42W ceiling.
Evidence kind: adapter-output
Limits: A replacement adapter shared across products, not measured gateway consumption.
Open the official source · checked 2026-08-04
TP-Link · M7000
M7000 Mobile Wi-Fi
Published value: 5V/1A external power, a 5W input ceiling.
Evidence kind: volts-amps-ceiling
Limits: One model's input ceiling, not measured in-use charging draw or a hotspot-class range.
Open the official source · checked 2026-08-04
ASUS · ZenScreen Go MB16AWP
ZenScreen Go MB16AWP Tech Specs
Published value: 7.6W power consumption; separate 5V/3A (15W) DC input rating.
Evidence kind: published-consumption
Limits: One 15.6-inch model, not a complete portable-monitor class range.
Open the official source · checked 2026-08-04
Signify / Philips · 929001998105 LED lamp
Philips LED lamp 929001998105
Published value: 8W consumption.
Evidence kind: published-consumption
Limits: One lamp supports the 8W exemplar, not a range spanning bulbs, lanterns, panels, and string lights.
Open the official source · checked 2026-08-04
Sony · BC-ZD1 dual-battery charger
Troubleshooting the BC-ZD1 battery charger
Published value: 30W (15V/2A) recommended; 18W works more slowly; 5V/1.5A is insufficient.
Evidence kind: adapter-input-requirement
Limits: Input requirement for one dual-battery charger, not measured wall draw for all camera chargers.
Open the official source · checked 2026-08-04
Valve · Steam Deck
Steam Deck Tech Specs
Published value: APU power 4–15W; bundled PD 3.0 supply 45W.
Evidence kind: component-power-and-adapter-output
Limits: Component power and adapter capacity are different quantities and do not equal whole-device wall draw.
Open the official source · checked 2026-08-04
Vornado · VFAN Jr.
VFAN Jr. Whole Room Vintage Circulator
Published value: 15W and 30W speed ratings.
Evidence kind: published-consumption
Limits: One compact circulator, not a complete small-fan class.
Open the official source · checked 2026-08-04
Vornado · OSCR37
OSCR37 37-inch Oscillating Tower Circulator
Published value: 33W, 39W, 45W, and 49W speed ratings.
Evidence kind: published-consumption
Limits: One tower circulator, not a complete small-fan class.
Open the official source · checked 2026-08-04
ResMed · AirSense 11
AirSense 11 User Guide
Published value: 56.1W typical and 73.2W peak under the guide's documented test; 65W power supply.
Evidence kind: published-consumption
Limits: One device/test condition; it does not isolate every accessory or establish requirements for another prescribed setup.
Open the official source · checked 2026-08-04
U.S. Environmental Protection Agency, ENERGY STAR · Compact and full-size refrigerator categories
Refrigerators & Freezers Key Product Criteria
Published value: Defines compact refrigerator/freezer as less than 7.75 ft³ and describes annual-energy test context.
Evidence kind: government-test-method-and-category
Limits: No running watts, compressor duty cycle, or startup-surge values.
Open the official source · checked 2026-08-04
Generac · Generic sump-pump sizing table
GP Series Portable Generator Owner's Manual
Published value: 800–1,050 running watts; allow three times listed watts for starting.
Evidence kind: generic-sizing-estimate
Limits: Manufacturer planning estimate, not laboratory data for a specific pump; cycle frequency is not addressed.
Open the official source · checked 2026-08-04
Starlink · Standard Actuated, Standard 4/4X/Enterprise, and Mini examples
How much power does my Starlink need?
Published value: AC-input averages including terminal, router, power supply, and cables: 50–75W; 75–100W; Mini 20–40W.
Evidence kind: manufacturer-published-range
Limits: Named Starlink generations only; values vary with temperature, location, and use.
Open the official source · checked 2026-08-04
Starlink · Starlink Mini
Starlink Mini Specifications
Published value: 25–40W average, 60W input rating, 100W USB-PD requirement.
Evidence kind: manufacturer-published-range
Limits: The 25–40W average differs from the support article's 20–40W range; both are disclosed.
Open the official source · checked 2026-08-04
Seagate · Specific bare 3.5-inch Exos drive
Exos recertified drive data sheet
Published value: 6.7W idle and 10.5W maximum random-read consumption.
Evidence kind: published-consumption
Limits: Bare-drive component consumption excludes an external enclosure, bridge electronics, and wall-adapter losses.
Open the official source · checked 2026-08-04
Samsung · UN32H5000FFXZA
32-inch H5000F Smart TV (UN32H5000FFXZA)
Published value: 38W typical, 50W maximum, and 0.5W standby.
Evidence kind: published-consumption
Limits: One 32-inch model, not a complete 24–32 inch television class.
Open the official source · checked 2026-08-04
BenQ · GV1
GV1 Portable Projector Data Sheet
Published value: 24W maximum, 16.7W normal, and 13.1W eco consumption.
Evidence kind: published-consumption
Limits: One low-power portable projector, not a class range.
Open the official source · checked 2026-08-04
BenQ · GV50
GV50 Specifications
Published value: 85W typical power consumption.
Evidence kind: published-consumption
Limits: One brighter portable projector, not a class range.
Open the official source · checked 2026-08-04
Correction log
Version 1.0.0 · August 4, 2026: added a source/evidence contract and downloadable artifacts. Widened camera-charger and gaming-laptop planning envelopes; revised the CPAP example around a published named-model test; widened the projector range; narrowed the sump-pump row to the cited generic 800–1,050W sizing range and labeled its 3× start allowance; scoped the common USB-C charger row to 18–100W; preserved the Starlink Mini 20–40W versus 25–40W source discrepancy.
Device IDs did not change, so existing calculator and shared-plan links remain compatible.
Turn a wattage into a plan
Some links on this page may be paid links. If you buy through them, Cynosure LLC may earn a commission at no extra cost to you. We do not claim to have personally tested products unless clearly stated.
Frequently asked questions
Are these wattages exact for my device?
No — most are source-informed editorial planning ranges, not manufacturer ranges for every model and not PowerPreflight measurements. Your device may sit anywhere in the range, or outside it. Check its label and manual, then measure representative use with a plug-in watt meter when the purchase depends on the result.
Should I size with the typical watts or the high end of the range?
Use the typical figure for an exploratory runtime estimate and lean toward the high end when sizing a battery you have not bought yet. Our calculators apply an 85% efficiency assumption and hold back a 10% reserve, but those factors do not cover an unknown device draw or startup surge. Verify both when the purchase depends on the result.
Do these figures include startup surge?
The main range is running watts. A Surge badge means startup can briefly exceed that range, but the multiplier is device-specific and must be verified. Only the generic sump-pump sizing row includes a source-based 3× planning allowance; refrigerator surge remains unlabeled numerically until model-specific evidence is available.
Calculations are estimates only. Real runtime depends on battery age, inverter efficiency, device behavior, temperature, surge loads, manufacturer limits, and actual measured wattage. Always verify product specifications before buying or relying on a setup.