On this page
The watt-hour number printed on the box is the first thing you look at and the first thing that misleads you. It’s not your runtime — it’s a ceiling before any real-world losses, and the gap between that ceiling and what you actually get is where most outage plans fall apart.
Here’s the shape of the real calculation: you start with rated capacity, lose roughly 10–15% to inverter conversion before a single device sees power, hold back some reserve to avoid deep-discharging the cells, then divide by your actual load. And that load is almost certainly wrong, because the watts printed on your fridge or sump pump are running watts — not the startup surge that can hit several times higher and trip the inverter entirely. The thing that ends an outage early isn’t a slow, steady drain to zero. It’s one compressor kicking on at the wrong moment and the inverter shutting down.
That’s the shape of what this guide untangles. Worth noting upfront: the detailed numbers in the research behind this piece trace almost entirely to one seller’s published material. Where that matters — and it matters in a few specific places — it’s called out directly.
The Runtime Math (and Why It Always Comes Out Wrong)
The published version of the runtime calculation looks clean: multiply each device’s watts by its hours of use, add them up, compare to your unit’s capacity. A fridge at 80W running for 24 hours, a few LED lights, a router, two phones — on paper that totals somewhere around 2,100Wh for a full day of basics.
The problem is everything the formula leaves out. Before any device sees power, the inverter takes its cut — roughly 10–15% gone before you’ve run anything. That same 2,100Wh demand now needs closer to 2,400Wh of stored capacity to actually cover it. On top of that, the unit’s own standby electronics — the BMS, the display, the circuitry that keeps it ready — draw power continuously even when nothing’s plugged in. Over a multi-day outage, that idle draw is a meaningful leak.
Then there’s the load estimate itself. The 80W fridge figure is an average running watt number — it doesn’t reflect the compressor cycling on from a warm state, or the startup surge when it kicks back on after a dormant period. The actual draw fluctuates, and the peak moments are what stress the inverter.
A working planning rule: take your unit’s rated Wh, apply roughly 85% to account for inverter overhead, then divide by your summed running loads. Treat the result as an optimistic estimate, not a guarantee — because cold ambient temperatures, hard-working appliances, and standby draw all pull the real number lower.
The “One Day” Benchmark Is Someone Selling You Something
You’ll see figures like “2,000Wh is the minimum for a full day of household backup” repeated across product pages. That framing is worth examining. The specific example — fridge, lighting, router, phones adding up to roughly 2,100Wh — comes from a seller whose flagship unit at time of writing is rated at 2,048Wh. The requirement conveniently matches the product.
That doesn’t make the planning figure useless. A frugal essentials load for one day really does land somewhere in that range, and it’s a reasonable starting point. But “minimum for one full day” quietly assumes two things that often don’t hold in real outages: that the outage ends in 24 hours, and that you never run anything with a heating element or a motor beyond the fridge. Space heater? Kettle? Microwave? Each of those blows past the estimate on its own.
Use the 2,000Wh ballpark as a floor for a disciplined essentials load, not a requirement derived from independent measurement. Size to your actual appliance list. And if there’s any chance the outage runs more than a day, multiply accordingly — without recharging, the math is strictly linear.
Motor Loads: The Surge Is the Spec That Actually Matters
This is where outage plans fail most decisively, and most product pages skip it entirely.
A sump pump running draws somewhere in the range of 600–1,000W. That figure tells you how fast it drains the battery. It does not tell you whether the pump will start at all. The moment a motor kicks on, it demands a brief but massive surge of current — commonly several times the running draw — to break inertia and spin up. That fraction-of-a-second spike is called the startup or locked-rotor surge, and if your inverter’s surge rating isn’t high enough to absorb it, the unit shuts off. The motor never starts. Your basement floods.
Running wattage determines your runtime estimate. Surge wattage determines whether the thing starts. Both numbers matter, and you need both from your appliance’s documentation — not just the label watts. This applies to fridges, well pumps, sump pumps, and any other compressor or motor-driven device.
A station with enough continuous output and enough surge headroom can handle short, intermittent pump runs during a flood event. It is not a substitute for extended continuous duty — frequent cycling will drain even a large unit quickly.
Solar Recharging: Read the Fine Print on Those Time Claims
Vendor recharge time claims are best-case numbers, and some of them don’t even survive the vendor’s own math.
One published example: a roughly 2,048Wh unit recharging in approximately 3 hours on a 220W panel in “good conditions.” Run the arithmetic: 220W for 3 hours yields around 660Wh, not 2,048Wh. The claim only makes sense if the unit started nearly full — making it a top-up time, not a full-recharge time, dressed up as the latter.
That kind of internal inconsistency is a useful reminder of how to read any solar recharge claim. Even setting aside the math, field output from a solar panel falls well short of its rated wattage under anything less than perfect conditions — clouds, haze, low sun angle, heat, dust, and panels that aren’t actively tracked all cut into it. During the kind of multi-day storm outage when you most need backup power, solar input can be close to zero for the duration.
Solar recharging is a genuine advantage for outages in good weather and a much less reliable one in bad weather. Plan around the pessimistic case, not the spec sheet best case.
Keeping It Ready: Storage and the Self-Discharge Problem
The quietest failure mode for a preparedness power station isn’t a bad battery or a blown inverter. It’s the unit that was fully charged when you bought it, stored in a garage, and never touched again — until the outage, when it won’t turn on.
Lithium units self-discharge over time. One seller cites a range of roughly 3–6 months before meaningful capacity loss, with the pace influenced by chemistry, temperature, and whether the unit’s own standby electronics are drawing power. Heat speeds both self-discharge and long-term cell degradation. A unit stored at full charge or near zero charge for extended periods also degrades faster than one kept in the middle range.
The practical maintenance routine:
- Check and top up charge every 2–3 months.
- For longer-term storage, aim for a 50–80% state of charge, not full and not empty.
- Store in a cool, dry location — a hot garage is among the worst options.
These figures come from a single manufacturer’s guidance and align with general lithium storage practice, but haven’t been independently verified against long-term testing. Treat them as reasonable directional guidelines, not certified maintenance intervals.
Cycle Life Claims: What the Number Doesn’t Tell You
LiFePO4 (LFP) chemistry has become the default in serious power stations partly because of its cycle life reputation — figures like “up to 3,000 cycles” appear routinely on spec sheets. LFP does meaningfully outlast older lithium chemistries under typical conditions. But the bare cycle number, as usually stated, is missing the piece that makes it useful.
Cycles to what? A battery at 40% of its original capacity technically still completed those cycles. The honest version of a cycle-life claim specifies the capacity retention threshold — typically 80% — at which point the unit is considered end-of-life for the test. Without that, the number is unverifiable marketing. No reviewer can cycle a unit 3,000 times within any realistic test window, and the missing condition isn’t an oversight — it’s where the claim’s teeth would be if they existed.
Use LFP cycle-life claims as a relative comparison (longer-lived than other chemistries, all else equal), not as a warranty or a lifespan guarantee. Heat and deep discharges degrade cells faster regardless of chemistry.
The Number That Actually Ends Your Outage
Everything above points to the same underlying principle: the figure on the box measures one thing, and your outage is determined by several others. Capacity gets you in the ballpark. Inverter efficiency, standby draw, and your real load profile close the gap between the spec and your actual runtime. And the startup surge of motor-driven appliances — the number that rarely appears on the product page — is the one most likely to end the game early, not with a slow drain but with a sudden shutoff.
Size to your real appliance list, not the marketing example. Budget for conversion losses. Know your appliances’ surge ratings, not just their running watts. And check the charge level before the outage hits, not during it.