Why a Microwave Trips the Breaker on Defrost

This page contains affiliate links. If you buy through one we may earn a commission, at no cost to you.

There are two completely different answers and the timing of the trip tells you which one you have. If it goes the instant you press start, and only sometimes, the cause is the surge that energises the high voltage transformer. If it goes a minute or several into cooking, and repeatably, the cause is the steady draw against the breaker's trip curve, usually with something else sharing the circuit.

The useful finding sits in the first case, and it is the opposite of what people try. On a conventional oven, a lower power setting does not reduce the current. It switches the magnetron on and off at full power, so ten minutes on a defrost setting energises that transformer twenty times instead of once, and makes a trip roughly thirteen to eighteen times more likely than running the same food at full power for a third of the time.

The number on the door is not the number at the wall

Start with what the oven actually draws, because the figure everyone quotes is the wrong one. "1000 watts" on a microwave is the microwave power delivered into the cavity. The wall has to supply that plus the magnetron's own losses, the transformer's losses, the cooling fan, the turntable motor and the controls.

One listing in our catalogue publishes both halves, and it does it in the most useful possible place. A Toshiba 1.2 cubic foot countertop microwave publishes 1000 watts of cooking power, and among its drawbacks states: "Input power is high at 1600 watts". That single line is the most informative thing in the category. It makes the efficiency 62.5 percent, and it makes the current 1600 divided by 120, which is 13.33 amps.

Hold that against the circuit. A 15 amp branch is the common kitchen case, and the ordinary derating for a load that runs a long while is 80 percent of the breaker, which is 12 amps. So 13.33 amps is 89 percent of the breaker and 111 percent of the continuous limit. The oven on its own is above the figure a circuit is supposed to carry indefinitely, and below the figure that opens the breaker, which is precisely the condition that produces intermittent and confusing behaviour.

A Toshiba 0.9 cubic foot 900 watt microwave is honest about the same thing without supplying the number, its drawbacks noting that "Input power is higher than output, indicating energy use". Apply the same 62.5 percent and it is 1,440 watts, which is 12.0 amps, sitting exactly on the continuous limit. One step smaller again, a Nostalgia retro 0.7 cubic foot 700 watt oven works out at 1,167 watts and 9.7 amps, comfortably under it. Its own drawback that "700 watts is lower than many standard microwaves" and "May take longer to heat food" is true, and the same property is the reason it is the one oven here that will never trip anything on steady draw.

A breaker does not trip at its rating

This is the part that makes the first failure mode so hard to recognise. A thermal magnetic breaker has a time current curve, and the time is as important as the current. At its rating it holds indefinitely, by design. At about a third over it takes in the region of an hour. At twice rating it is around a hundred seconds. At three times, under ten. And above eight to ten times rating a separate magnetic element opens it inside a single half cycle.

So 13.33 amps on a 15 amp breaker never trips it, no matter how long you cook. What trips it is the second load. Our own page on whether kettle wattage really matters works out that a 1500 watt kettle pulls 12.5 amps and that this is already 83 percent of a 15 amp circuit, which is correct and is the other half of this arithmetic. Put the two on one circuit and you have 25.83 amps, which is 1.72 times rating, which lands between the hour and the hundred seconds. A trip a minute or two into reheating, while the kettle is on, with no trip when either runs alone. That is the whole pattern, and it is a circuit loading problem rather than an appliance fault.

The surge, and why it is random

Now the other failure, the one that goes at the moment of starting. A conventional microwave contains a mains frequency high voltage transformer, and energising any transformer draws a magnetising surge whose size depends on where in the AC cycle the contacts happen to close and on how much magnetism was left in the iron core when it was last switched off. Both of those are effectively random from one start to the next.

Put numbers on it. At 13.33 amps RMS the ordinary peak is 18.85 amps. Magnetising inrush on a small transformer energised at an unfavourable instant runs eight to twelve times the rated peak, so 151 to 226 amps for the first half cycle. The instantaneous element on a residential 15 amp breaker sits at eight to ten times its rating, which is 120 to 150 amps.

Those two ranges overlap. Not every start, not predictably, but some fraction of starts. Which is exactly the reported symptom: it trips sometimes, on the same oven, in the same socket, doing the same thing, and the owner cannot find a pattern because there is not one to find.

Worth adding that the supply inside the oven is not a tidy sinusoidal load even when it is running. Our page on what interferes with a baby monitor gets this right from a radio point of view, noting that "the magnetron inside it is driven from the mains, so it pulses on and off at the mains frequency". The same half wave supply that makes a microwave pulse its radio output at 60 hertz also makes it draw current in pulses, so its peak current is higher relative to its RMS current than a resistive load of the same wattage. That eats into the margin before the magnetic element, which is why two appliances of identical wattage do not stress a breaker identically.

Why the gentle setting is the dangerous one

Here is the finding. A conventional microwave has one magnetron power and no way of turning it down. Every setting below maximum is produced by switching the magnetron on and off on a fixed period, typically twenty to thirty seconds long. Power level three means it runs full power for about nine seconds out of every thirty and sits dark for the other twenty one.

Which means the number of transformer energisations is not one. It is the duration divided by the period. Ten minutes of defrost at a thirty second period is twenty separate energisations, each one an independent roll of the dice described above.

If a single energisation carries a five percent chance of landing in the overlap, the chance of getting through twenty of them is 0.95 to the twentieth power, which is 35.8 percent, so the chance of a trip is 64.2. Against 5 percent for a single full power start. At two percent per energisation the figures are 33.2 against 2. At one percent, 18.2 against 1. The ratio is thirteen to eighteen times whichever number you pick, because it is the compounding that does the work rather than the probability.

So the setting chosen out of caution is the one that trips the breaker, the setting feared as too powerful is the safe one, and nothing in the kitchen gives any hint of this, because the two settings draw identical current while the magnetron is on.

The exception, which our own roundup already describes correctly

There is one design that does not do this. A Panasonic NN-SN65KW with inverter technology publishes 1200 watts and 11 power levels, and in an inverter oven those levels are real. A switching supply varies the power delivered to the magnetron continuously rather than chopping the whole appliance on and off, so the transformer, or rather what replaces it, is energised once at the start of the run and stays energised.

One energisation for a twenty minute defrost instead of forty. Our countertop microwave roundup has the mechanism exactly right already, saying that "Standard microwaves switch full power on and off to lower the setting, which can overcook edges" while "Inverter models deliver steady lower power for gentler, more even reheating". Every word of that is correct. What it frames as a cooking benefit is also, and nobody mentions this anywhere, the only genuine fix on this page for a breaker that trips at the moment of starting.

What not to do, and what the breaker is for

One thing is worth stating plainly before the fixes. The breaker is not protecting the microwave. It is protecting the wire in the wall, and the wire sets the rating rather than the other way round. Ordinary 14 gauge copper is rated 15 amps and 12 gauge is rated 20. Fitting a larger breaker to stop the nuisance leaves the same wire behind a device that will now let more current through it than it was sized for, which moves the problem from an inconvenience to the wiring. A breaker that keeps operating is reporting something, and the thing it is reporting is in the two sections above.

The honest limit

The weak number here is the probability per energisation, and it is weak because nothing publishes what it needs. No microwave listing we carry states a transformer rating, an inrush current, or a duty cycle period, so the count of energisations for a given setting cannot be read off a listing and has to be inferred from a typical twenty to thirty second cycle. And the instantaneous trip threshold is not marked on any breaker either: eight to ten times rating is the usual design range, and the actual figure varies by maker.

So the five percent is illustrative and the sixty four percent that follows from it is illustrative too. What is not illustrative is the structure. One energisation becomes twenty, the probability compounds as one minus the survival probability to the twentieth power, and that ratio holds for any per start probability you care to assume. The steady draw half is firmer still, because 1600 watts is a published figure and 13.33 amps is division.

The technique that needs nothing

Three moves, and the third is the interesting one.

First, diagnose by timing rather than by guessing. Instant and intermittent is inrush. Minutes and repeatable is loading. They have different fixes and no overlap.

Second, for the loading case, stop sharing the circuit. The kettle alone is 12.5 amps and the oven alone is 13.33, and either is fine. Together they are 172 percent of the breaker. Nothing needs replacing.

Third, for the inrush case, run full power for a shorter time and then let the food stand, which replaces the low power setting entirely. The same total energy arrives, and the reason a low setting exists at all is evenness rather than gentleness. Standing time buys the same evenness for nothing, and the right duration is computable.

Microwave hot and cold spots are not random. They are the standing wave inside the food. At 2.45 gigahertz the free space wavelength is 122.4 mm, and inside water rich food, whose relative permittivity is in the region of 60, that shrinks by the square root of 60 to 15.8 mm. So hot and cold regions sit about 7.9 mm apart. Evening out across a quarter of that wavelength, 3.95 mm, by conduction alone takes the distance squared over the thermal diffusivity, and at the 1.3 times ten to the minus seven square metres per second that water rich food runs at, that is 120 seconds.

Two minutes. The standing time printed on frozen food packets, which everyone ignores, is the diffusion time across half a microwave wavelength, and it does the job that twenty magnetron energisations were being used to do. The same arithmetic appears in our page on why an air fryer browns the crust before the middle cooks, where the distances are larger and the times correspondingly longer, because conduction through food is a square law wherever you meet it.