Watts vs Watt-Hours: Why the Number on the Box Misleads You
Figures last verified
Watts measure rate. Watt-hours measure quantity. A power station's watt rating tells you what it can run at any instant, and its watt-hour rating tells you for how long. A product called a 1000 might have either figure attached to that number, and buying the wrong one is how people end up with a unit that starts their refrigerator and dies in three hours, or one that would last two days if only it could start the compressor.
The distinction in one analogy
A garden hose has a flow rate and a bucket has a volume. Watts are the flow rate. Watt-hours are the volume of the bucket.
A wide hose fills a small bucket quickly and then you are out of water. A narrow hose attached to a swimming pool never runs dry but cannot fill a watering can in any useful time. Backup power products come in both shapes, and the marketing rarely makes clear which one you are looking at.
Formally, one watt-hour is one watt sustained for one hour. A 100 watt load running for ten hours consumes 1,000 watt-hours. The same 1,000 watt-hours could be spent in one hour by a 1,000 watt load, or in six minutes by a 10,000 watt load, if anything could deliver it that fast.
Which number is which on real products
Portable power stations are the worst offenders because manufacturers name models after whichever number sounds better.
The watt-hour figure is capacity. It appears as Wh or kWh, sometimes buried in a specification table rather than on the front of the box. This is the size of the battery. A unit listed at 1,024Wh holds roughly a kilowatt-hour of energy.
The watt figure is output. It appears as W, usually split into continuous or rated output and a separate peak or surge output. This is what the inverter can deliver at once. A unit listed at 1,000W continuous and 2,000W peak can sustain a kilowatt indefinitely until the battery empties, and can briefly handle twice that.
A product marketed as a 1000 might be 1,024Wh with a 1,000W inverter, which is coherent. It might also be 1,000W of output attached to a 600Wh battery, or 1,000Wh attached to a 300W inverter. Both of those exist. The only way to know is to read the specification table.
Generators are less ambiguous, because they have no battery at all. A generator's rating is always in watts, both running and starting. Its equivalent of capacity is fuel, which is why the useful question there is consumption per hour rather than watt-hours, worked through in how much fuel a generator uses during a three-day outage.
Home batteries are usually honest about both, listing kWh capacity and kW continuous output as separate headline figures.
The two questions, asked properly
Every backup power decision is two independent checks, and passing one does not imply passing the other.
The output check asks whether the load can run at all. Add up the running watts of everything on at once, find the largest single surge, and confirm the inverter's continuous rating exceeds the first and its peak rating exceeds the sum. Figures for common appliances are in the appliance surge and running watts table. Fail this check and nothing happens. The inverter trips, the display shows an overload fault, and battery size is irrelevant.
The capacity check asks for how long. Add up the watt-hours per day of everything you plan to run, derate for inverter losses, and compare against the battery's usable capacity. Daily consumption comes off the EnergyGuide label. Fail this check and everything works beautifully until it stops.
These fail in different ways and are fixed by different purchases. An output failure is fixed by a bigger inverter or a soft starter. A capacity failure is fixed by a bigger battery or by recharging, which for most people means solar and leads to how many panels it takes to recharge in a day.
Where the arithmetic goes wrong
The most common error is multiplying the wrong pair of numbers. A 1,000Wh station running a 100W load does not last ten hours in practice. Inverter efficiency of roughly 85 to 90 percent takes a cut, and small loads are where inverters are least efficient because the inverter's own idle consumption becomes significant relative to the load. Running a 10W router off a large station can waste more energy on inverter overhead than on the router.
The second error is using running watts to estimate runtime for a duty-cycling appliance. A refrigerator listed at 200 running watts does not consume 200 watt-hours per hour, because the compressor is off most of the time. Actual consumption averages somewhere near a third of that. This is why label-based watt-hour figures beat wattage-based estimates for anything with a thermostat, and why runtime predictions built on running watts come out drastically short. The real numbers are in how long a power station runs a refrigerator or freezer.
The third error is treating amp-hours as capacity without checking voltage. A 100Ah battery is meaningless until you know its voltage. At 12V nominal it holds 1,200Wh. At 24V the same amp-hour figure holds 2,400Wh. Amp-hours multiplied by nominal voltage gives watt-hours, and comparing amp-hour ratings across different voltages compares nothing at all.
The fourth is ignoring usable versus nameplate capacity. Lead acid chemistry should not be discharged below roughly half its capacity without shortening its life dramatically, so a 100Ah lead acid battery offers about 600Wh in practice at 12V. Lithium iron phosphate tolerates deep discharge far better and delivers most of its nameplate. That difference is larger than the price gap suggests and is covered in LiFePO4 versus lead acid.
Solar uses both units too
Panels are rated in watts, and that rating is instantaneous output under standard test conditions. A 400W panel produces 400W at the moment conditions match the laboratory, which they essentially never do.
What you actually harvest is watt-hours per day, and it depends on how many hours of useful sun the panel sees, its angle, its temperature, and the efficiency of the charge controller. The rule of thumb is that a panel yields its rated wattage multiplied by roughly four to five peak sun hours in good summer conditions, and considerably less in winter or under cloud. A 400W panel might return 1,200 to 1,600Wh on a good day and a few hundred on a poor one. Why the gap between rating and reality is so wide is the subject of why solar panels never hit their rated wattage.
Charging is also subject to an output limit. A station with a 200W maximum solar input cannot accept more than 200W no matter how many panels you attach, so oversizing the array beyond that ceiling buys you better performance in poor light but no more peak throughput.
A worked example
Suppose you want to run a refrigerator, a chest freezer, a router, and some LED lighting through a two-day outage.
Output check. Refrigerator running at 150W, freezer at 150W, router at 20W, lighting at 40W gives 360W continuous. The largest single surge is the freezer compressor at perhaps 1,200W. So you need an inverter rated above 360W continuous and above roughly 1,560W peak. A unit advertised at 1,000W continuous and 2,000W peak passes.
Capacity check. The refrigerator's label says 400 kWh per year, so 1,100Wh per day. The freezer's says 350 kWh per year, so about 960Wh per day. Router and lights add perhaps 400Wh. That is roughly 2,460Wh per day, or 2,900Wh after inverter losses, and about 5,800Wh across two days.
The same unit that passed the output check at 1,024Wh fails the capacity check by a factor of five. You need either six times the battery, or a way to put roughly 3,000Wh back in every day, which is where solar or a generator enters the plan. The comparison across those three approaches is in which backup actually fits your house.
That mismatch is typical. Portable power stations are generally sized to pass output checks comfortably and capacity checks poorly, because inverters are cheap and batteries are not. Reading both numbers before buying is the entire defense.