ElectrifiedExperts
Buying a portable power station on rated watt-hours is like buying a car on fuel tank volume. Independent bench testing puts delivered energy anywhere from 68% to 93% of the number on the box, and no manufacturer publishes where their unit lands. Two more specs decide the purchase and are equally missing: the capacity endpoint a cycle rating was measured to, and how much the thing drains sitting in a closet doing nothing. All three below, plus the 0 °C charging wall.
The short version
- Rated watt-hours are not delivered watt-hours. Bench measurements span 68% to 93% of rated capacity, and no spec sheet tells you where a unit sits.
- “4,000 cycles” from Jackery and from EcoFlow are different tests. Jackery measures to 70% of original capacity, EcoFlow and Anker to 80%. DJI uses both inside one family.
- Not one manufacturer publishes standby drain. At 20 W idle a 1,000 Wh unit empties itself in two days with nothing plugged in. For emergency storage that is the decisive spec, and it does not exist.
- Almost every unit refuses to charge below 0 °C. Panels producing, sun up, state of charge frozen. The BMS is preventing lithium plating, and there is no override.
- “Solar generator” is a marketing phrase. Battery plus inverter plus MPPT in one case, and most listings using the term ship with no panel.
- X-Boost and Power Lifting are not a surge budget for motors. They drop output voltage, which is the wrong thing to do to an induction motor.
Number one: usable capacity, which nobody discloses
Every listing leads with watt-hours: nominal cell voltage times nominal amp-hours. That is not what reaches the AC outlet.
A 25-point spread. EcoFlow’s Delta Pro 3 returns 3,790 Wh of its 4,096 Wh rating. Anker’s SOLIX C1000 returns 860 Wh of 1,056. Small units do worse, because a small pack carries proportionally larger inverter overhead. Three separate losses hide inside that one percentage.
Inverter conversion loss. The pack is DC, your appliance wants 120 V AC. A pure sine inverter well inside its rating runs high 80s to low 90s percent, and much worse at light load, because its electronics draw a fixed overhead whatever you plug in. A 40 W laptop on an 1,800 W inverter is a bad operating point.
BMS low-voltage reserve. The battery management system stops the discharge above the cells’ true floor. LiFePO4 sits flat near 3.2 V per cell then falls off a cliff, so designers cut off early to stay clear of it. That reserve is energy you will never see.
Nominal versus actual pack voltage. Jackery rates the Explorer 1000 v2 at 1,070 Wh from 30.4 Ah at 35.2 V nominal. The pack does not sit at 35.2 V through a discharge. Average voltage under load runs below nominal, so real pack energy is under the arithmetic product before anything downstream touches it.
Separating them means discharging one unit twice at matched power, through the 12 V DC port and through the AC outlet. DC out over rated gives pack plus BMS reserve; AC out over DC out gives true inverter efficiency. Nobody publishes this.
Number two: cycle life measured to endpoints that do not match
A cycle rating means nothing without its capacity endpoint. “4,000 cycles” means “after 4,000 full cycles the pack still holds at least X% of original capacity.” Change X and you changed the test.
Jackery rates the Explorer 1000 v2, Explorer 2000 Plus and HomePower 3000 at 4,000 cycles to 70%. EcoFlow rates the Delta 3 Plus and Delta Pro 3 at 4,000 to 80%, the Delta 2 and River 3 Plus at 3,000 to 80%, the Delta Pro Ultra at 3,500 to 80%. Anker rates the C1000 Gen 2 at 4,000 to 80%. Degradation is roughly monotonic, so reaching 70% takes substantially more cycling than reaching 80%. Jackery’s number is the harder one.
That this is a specification problem rather than a brand-quality problem is clearest at DJI. The Power 1000 is rated above 70% after 4,000 cycles, the Power 2000 above 80% after 4,000 cycles. Same maker, same generation, two endpoints, both printed as “4,000 cycles”.
Anker’s F3800 goes further with “3000+ cycles” and no endpoint at all. That is a number, not a spec.
Number three: standby drain, and the industry’s silence
Here is the scenario that sells most of these units. You charge one to 100% for hurricane season, put it in a closet, and expect it to work in March. That depends entirely on parasitic draw. We checked Jackery, EcoFlow, Anker, Bluetti, Goal Zero and DJI. Not one publishes it.
Those bars are arithmetic on a 1,000 Wh pack, not measurements, because public measurements do not exist. Look at the shape anyway. At 5 W the unit survives eight days, at 20 W two. That is the gap between a working emergency kit and a dead brick.
Victron, a component maker with no consumer marketing to protect, publishes self-consumption for its SmartSolar MPPT controllers down to the milliamp: 30 mA at 12 V, or 0.36 W. If a charge controller can state it, so can a power station. Every unit here also keeps a Wi-Fi or Bluetooth radio alive for its app, a continuous load nobody has quantified against the radio-off case. Until real numbers exist, store with the inverter off and top up quarterly.
The 0 °C charging wall
Read the temperature line on almost any spec sheet here and you find the same pair of ranges: charge 0 to 45 °C, discharge −10 to 45 °C. Verified on the Jackery Explorer 1000 v2 and 2000 Plus, EcoFlow’s Delta 3 Plus, Delta 3 Max Plus, Delta Pro 3 and River 3 Plus, and the DJI Power 1000. Anker’s C1000 Gen 2 states 32 to 104 °F, the same floor in other units.
The consequence catches people out every winter. On a clear freezing morning the panels are producing, the unit shows solar connected, and the state of charge does not move. Nothing is broken. Charging a lithium cell below freezing plates metallic lithium on the anode, permanently and eventually dangerously, so the BMS refuses. Discharge is friendlier: you can pull power to −10 °C at reduced capacity. A unit in an unheated garage runs your lights all night, then refuses the sunrise.
The exception worth naming is EcoFlow’s Delta Pro Ultra, which claims −20 to 45 °C operating. Self-heating LiFePO4 packs solve it properly in the DIY world, warming the cells before accepting charge. No mainstream portable power station does this, so cold-climate buyers charge indoors or accept that solar is a spring-to-autumn feature.
“Solar generator” is a marketing term
Nothing is generated. The product is three components in one case: a LiFePO4 battery, a pure sine inverter, an MPPT charge controller. The word captures search intent from people typing “generator” after an outage, and it sidesteps the comparison “battery” invites. A 2 kW engine generator with a jerry can beside it runs as long as you keep feeding it. A 2 kWh power station runs until it is empty.
The sharpest tell: Amazon titles for the Jackery Explorer 1000 v2, the Bluetti AC70 and AC180 and the Anker C800X all say “Solar Generator” while the listing ships with no panel. The two product classes are also certified under different safety standards, so the shared noun carries no shared meaning.
X-Boost and Power Lifting will not run your well pump
EcoFlow’s X-Boost (2,200 W on the Delta 3 Plus, 6,000 W on the Delta Pro 3) and Bluetti’s Power Lifting (3,600 W on the AC200L, 3,900 W on the Elite 200 V2) let a unit run appliances rated above its continuous inverter output. They do it by reducing output voltage. A resistive load, a kettle or a heater element, then draws its rated current at lower voltage and lower power. It heats more slowly, and nothing is harmed. Bluetti says so in its own copy: the Elite 200 V2’s 3,900 W figure is specified for pure resistive loads. That qualifier is the whole story and it is almost never quoted.
An induction motor is not a resistive load. A well pump or a fridge compressor is roughly a constant-power device. Feed it lower voltage and it draws higher current for the same mechanical output, and more current in the windings means more heat. If voltage sags far enough that the motor cannot develop breakaway torque, it sits at locked-rotor draw until a thermal overload trips, and repeated stalls cook winding insulation.
So when a comparison table says a 2,400 W unit runs a 3,600 W well pump because Power Lifting exists, that advice can destroy the pump. Size motor loads on rated continuous and rated surge, and pretend the boost figure is not there. The credible 240 V motor-load machines here are the EcoFlow Delta Pro 3, the Anker SOLIX F3800 and the Delta Pro Ultra. Jackery’s HomePower 3000 cannot, because despite the home-backup name it has no 240 V output.
Safety note: the EcoFlow Delta Max 2000 recall
On October 9, 2025 the CPSC announced recall 26-010 covering the EcoFlow Delta Max 2000, model EFD310: roughly 25,030 US units sold between July 2022 and May 2025, hazard that units may overheat and ignite, six fires and more than $850,000 in property damage. The remedy is a free firmware update, and owners were told to stop using the product until it is applied.
If you are shopping used or open-box, that is the model number to check and the firmware version to verify. A firmware-remedy recall also says something structural about the category: these are software-controlled battery systems whose safety envelope is enforced in code.
The units that survive a spec-sheet reading
| Model | Rated capacity | Cycles (endpoint) | AC output | Solar input | Notable |
|---|---|---|---|---|---|
| Jackery Explorer 1000 v2 | 1,070 Wh (also ships as 1,050 Wh) | 4,000 to 70% | 1,500 W / 3,000 W surge | 400 W combined, 2 inputs | Wall recharge in 1.58 h, 23.8 lb. Two capacities under one model name. |
| EcoFlow Delta 3 Plus | 1,024 Wh | 4,000 to 80% | 1,800 W / 3,600 W surge | “11V-60V 500W Max*2”, ambiguous | 13 ports, 2 × 140 W USB-C, under 10 ms UPS, expands to 5 kWh. |
| Anker SOLIX C1000 Gen 2 | 1,024 Wh (Gen 1 was 1,056 Wh) | 4,000 to 80% | 2,000 W / 3,000 W peak | 600 W at up to 60 V | 49-minute recharge, 20 dB. Two AC outlets; the USB-A pair share 12 W. |
| EcoFlow Delta Pro 3 | 4,096 Wh | 4,000 to 80% | 4,000 W / 8,000 W, 120/240 V | 2,600 W, two trackers | Measured 93% usable, best in the field. IP65, expands to 48 kWh, 113 lb. |
| Anker SOLIX F3800 | 3,840 Wh | “3000+”, no endpoint | 6,000 W / 9,000 W, 120/240 V | 2,400 W (3,200 W on the Plus) | NEMA 14-50 and L14-30, so it drives a real Level 2 EVSE. |
Capacities are nameplate figures, not delivered energy. Cycle endpoints are quoted as published; where a brand states none, that is shown rather than assumed.
In the 1 kWh class those three are close enough that secondary features decide it. Jackery is lightest, EcoFlow has the best ports, Anker recharges fastest but gives you two AC outlets.
EcoFlow DELTA 3 Plus
The best-rounded 1 kWh unit on Amazon: 1,024 Wh, 1,800 W continuous, 13 ports including two 140 W USB-C, sub-10 ms switchover, expandable to 5 kWh on packs shared with the older Delta 2 line. Its solar input copy reads “11V-60V 500W Max*2”, ambiguous between 500 W and 1,000 W.
Check price on Amazon(paid link)
Whole-home backup on 240 V circuits is a different purchase: dual-voltage output, a NEMA 14-50 or L14-30 outlet, expansion. Anker’s F3800 is the most outlet-complete unit in that class, listed on Amazon bundled with a 400 W panel. Its gap is longevity: “3000+ cycles” with no capacity endpoint stated.
Smaller is often the right call. The Jackery Explorer 1000 v2 bare unit on Amazon is the lightest 1 kWh option, and the Anker SOLIX C1000 Gen 1 listing on Amazon is the generation with the larger 1,056 Wh pack and six AC outlets, worth knowing before you assume newer means bigger. Bluetti’s AC180 on Amazon has the biggest nameplate in the class at 1,152 Wh and measured 82% usable.
What it does well
- LiFePO4 across the mainstream lineup, so thousands of cycles rather than hundreds
- Sub-20 ms switchover on most units, fine for a desktop or a router
- Wall recharge has collapsed to 45 to 60 minutes for 80%
- Quiet at low load, several units under 30 dB
Where it falls short
- Delivered energy runs 7 to 32% below rated and nobody says which
- Cycle ratings use different endpoints across brands, and inside brands
- Standby drain is unpublished industry-wide
- Charging stops below 0 °C on nearly every model, no override
- Boost modes are advertised in a way that invites motor loads
- Variant SKUs are a mess, and capacity changes without the model name changing
Frequently asked questions
How many watt-hours do I actually get from a portable power station?
Between about 68% and 93% of the rated figure at the AC outlet, on the only independent bench data set for this category. Larger units do better, because inverter overhead is a smaller share of a bigger pack. Assume 80 to 85%.
Why won't my power station charge in cold weather?
Because the battery management system blocks charging below 0 °C to prevent lithium plating on the cell anodes. Nearly every unit shares that limit. Discharging still works to around −10 °C, so a unit in a cold garage runs loads overnight, then refuses solar at sunrise.
Is 4,000 cycles better than 3,000 cycles?
Only if both were measured to the same capacity endpoint, and often they were not. Jackery measures to 70% of original capacity while EcoFlow and Anker measure to 80%, so Jackery’s 4,000-cycle rating describes a longer test. DJI uses both endpoints inside one family.
Can I run a well pump with X-Boost or Power Lifting?
No, and trying can damage the pump. Those modes lower output voltage so a resistive load draws less power, and Bluetti’s own spec for the Elite 200 V2 says the 3,900 W figure applies to pure resistive loads. A motor responds to lower voltage by drawing more current, risking stall and winding damage. Size on rated continuous and surge output only.
Will a power station hold its charge in storage for six months?
Nobody can tell you, because no manufacturer publishes standby draw. At 5 W idle a 1,000 Wh unit lasts roughly eight days, at 20 W roughly two. Store it with the inverter off and set a quarterly reminder to top it up.
Where to go next
If you have narrowed it to the big three, Jackery vs EcoFlow vs Anker works through which brand suits which buyer, including the SKU traps in each lineup. For the two units that can genuinely do home backup, there are spec teardowns of the EcoFlow Delta Pro 3 and the Anker SOLIX F3800. And if you bought one hoping to top up an electric car, read charging an EV from a power station first.
