How many watts does a 1,000 W microwave actually use?
Power basics Updated 2026-08-08
More than the number on the door, by roughly a quarter to a half. That figure is COOKING OUTPUT — how much microwave energy reaches the food — and the wall sees the input, which is larger by whatever the magnetron and its transformer lose. The record in this library draws 1,480 W at the socket.
The second surprise is the power-level dial. A conventional microwave has one output level and fakes the rest by switching the magnetron on and off, so a lower setting is a duty cycle rather than a smaller draw. The peak the inverter has to carry is the same on every setting.
Two watt figures, and the label shows the smaller one
What two makers state for the same appliance
Both figures are rating-plate input power, taken from the makers’ own manuals rather than from a review. Neither is the number on the door, and the gap between input and output is not a fixed percentage — it depends on the machine rather than on its size.
The rating plate on a countertop machine, which is the figure the door does not show. This is what the inverter carries while the magnetron is conducting.
A different brand, a LOWER cooking output, and a higher input figure. Whatever else the number on the door tells you, it does not rank machines by what they draw.
The empty row is why this site states no verdict for this appliance, and it is the honest reason rather than an oversight.
A high-voltage transformer has a real switch-on inrush. We have found no published figure for it across the Panasonic, Toshiba and GE documents this record cites, and this library declines to invent a multiple for a load class where the invented number would decide the answer.
Full power on a conventional machine: the magnetron conducts continuously, so the draw above is also the average.
The appliance maker’s own words for what the dial does: microwave energy seven-tenths of the time. Same peak, less of it.
And the same again at the bottom of the dial. Defrosting does not ask less of an inverter; it asks the same thing less often.
A load in the same room that this site does publish verdicts for, because it is purely resistive and has nothing to inrush.
See the record →Official specverified 2026-08-01microwave oven record · US 120 V
The ratio between the two figures is the part that surprises people who look it up carefully. Our record’s primary source states 1,480 W in for 1,200 W of cooking power, and its cross-check states 1,550 W in for 1,100 W out. That is 1.23× on one machine and 1.41× on the other. A rule of thumb that adds a fixed percentage to the door figure will be wrong on one of them.
A power level is a duty cycle
This is the best household demonstration of the quantity every runtime on this site is computed from. A conventional microwave cannot produce 700 W of cooking energy — it produces its one output level, or nothing, and alternates between the two. The appliance maker cited on this record puts it plainly: power level seven is microwave energy seventy per cent of the time. So the peak an inverter has to carry is identical at every setting, and only the watt-hours change: 1,480 W for the whole of a High cycle, and the same 1,480 W for 0.3 of a defrost cycle.
Why this site publishes no verdict for it
Set that against the kettle in the table above. It draws 1,500 W on the same kind of circuit, which is more than the microwave, and this site ranks every machine against it without hesitation — because a heating element has no inrush to find. The difference between the two is not size and not danger. It is that one of them has a number we could not find, and it happens to be the number that decides the answer.
What to do with your own microwave
Three steps, and the first one is the one most people skip.
The input figure is on the rating label at the back or inside the door frame, usually beside the current draw. On a US machine it will be stated at 120 V.
Since neither the makers nor this library can tell you the inrush, the honest approach is headroom on the continuous rating rather than a number that looks precise.
Open the source →Two minutes on High and four minutes at half power cost about the same energy, and ask the inverter for exactly the same peak.
Open the source →