Could a Microwave Oven Fool the Parkes Radio Telescope?

Could a microwave oven, opened a few seconds early in a staff kitchen a hundred metres away, make the Parkes radio telescope record what looked like a flash from across the universe?

The door opens while the oven is still running. The magnetron inside shuts down, and the receiver in the cabin above the 64 metre dish records a sweep near 1.4 gigahertz, high frequencies first, low ones trailing, the curve a burst from deep space would draw. Look at the sweeps in the 2015 paper that worked it out. If you had been handed one of those plots before the test, what would have made you look in the kitchen?

The signal beside the signal

In December 2014 CSIRO installed a new interference monitor at Parkes, a receiver on a rotating mount listening from 402 megahertz to 3 gigahertz, far wider than the telescope's own band. A month later it earned its place.

In the week of 19 January 2015 the telescope caught three perytons, on the Monday, the Wednesday and the Thursday. Each swept across the band at 1.4 gigahertz. And each time, the monitor caught something the dish did not: emission between 2.3 and 2.5 gigahertz, at the same moment, to within one time sample. The Giant Metrewave Radio Telescope in India was watching the same field for the third one. It saw nothing.

Which put the source somewhere on the Parkes site.

The band between 2.3 and 2.5 gigahertz is where microwave ovens work. There were three in frequent use near the dish: in the tower below the telescope, in the visitors centre, and in the staff kitchen in the building long known as the Woolshed, 100 metres from the dish. Two more stood at the observers' quarters about a kilometre away.

Emily Petroff wrote later, in The Conversation, that the first tests did not settle it. Then, "on March 17, almost two months after our initial find, we tested the microwave ovens in a slightly unusual way." Instead of letting the oven finish its cycle, they opened the door. "We tried stopping the microwave oven by opening the door and boom!"

The staff kitchen oven, with a ceramic mug of water inside, made three bright perytons, each at the moment the door opened. The visitors centre oven made six more. Both ovens were Matsushita, sold as National, and both were more than 27 years old. The trick worked about half the time.

The paper, by Petroff and fourteen colleagues, appeared in Monthly Notices of the Royal Astronomical Society in 2015. Its title was "Identifying the source of perytons at the Parkes radio telescope."

A lookalike in the archive

Perytons had been a problem for five years by then, and a doubt about the sky for most of that time.

In a paper posted in 2010 and published in 2011, Sarah Burke-Spolaor and colleagues searched 1,078 hours of Parkes data from 1998 to 2003 and found sixteen of them. They named them after the peryton, a winged elk that casts the shadow of a man, a creature from The Book of Imaginary Beings by Jorge Luis Borges and Margarita Guerrero. Eleven had come in a single 4.4 minute observation in 1998. All arrived in daylight, mostly mid-morning. All lit up all thirteen of the receiver's beams at once.

And their sweeps looked like the Lorimer burst. The burst found in the Parkes archive in 2007, the founding event of the fast radio bursts, had taken 355 milliseconds to cross the band. The perytons clustered around 360. In the units astronomers use, the dispersion measure, the burst stood at 375 and the perytons near 400.

Burke-Spolaor's team wrote that the new detections "cast doubt on the extragalactic interpretation of the original burst." Kocz and colleagues found five more in 2012 and noted pulses spaced about 22 seconds apart, a sign, they wrote, that the signals "may not be from a natural source." Bagchi, Cortes Nieves and McLaughlin found four more and no link to rain or lightning. Petroff's paper later counted 46 at Parkes since 1998.

The doubt went public. At a conference in 2011, Duncan Lorimer and Maura McLaughlin wrote in Scientific American, there was a show of hands on whether the Lorimer burst was real, and by then, they wrote, "most astronomers began suspecting that the Lorimer burst was a peryton, too." McLaughlin's own 2012 paper, with Bagchi, called the burst "unlikely" to belong to a cosmological population, "a conclusion that feels embarrassing now," they wrote.

The case was simple to state. If the telescope could make a deep-space sweep out of something on the ground, then the one burst it had found might be the same thing.

If the kitchen could fake the sky

Set aside for a moment what the ovens did. Ask the larger question. What does it mean if a kitchen can draw the same curve as the universe?

It would mean the first fast radio burst might have been lunch. A field that rested on one flash would have rested on a door opened early, and every paper that read the Lorimer burst as a flash from a billion light years away would have been reading a staff room.

It would mean that the instrument cannot, on its own, tell near from far. A dispersion sweep was the signature of distance, the one thing a flash could carry that said how much space it had crossed. If a magnetron winding down can draw it, the signature is not enough, and every single-dish detection of a single flash is a question about the building next door.

It would mean something larger again for any search for other minds. If our own machines can write the curve of the sky into our own instruments, then the first real signal from somewhere else could arrive looking exactly like our noise, and be filed with it. Our species would be listening for a stranger through a room full of its own echoes. On that reading the Wow! Signal, 72 seconds recorded by Big Ear in Ohio in 1977 and not found again in later searches, belongs to the same question: was it from there, or from here? A candidate that comes once, on one instrument, could be either, and the instrument alone could not say.

And that consequence is still in front of us. Every radio telescope on Earth listens from inside a civilization that leaks radio. On that reading the question of how much of what we record is us would have no final answer, only better monitors.

The standard account

The standard account holds that perytons are microwave ovens and the Lorimer burst is not.

Petroff's team set out why in four steps. The burst was strong in three beams, the pattern of a source in the sky the dish was pointing at; perytons lit all thirteen. The dish was pointing almost due south, and from that angle "the line of sight from the Woolshed microwave to the focus cabin is completely blocked by the telescope surface." The visitors centre oven had a line of sight, but its perytons came out near 200, not 375. And the burst arrived at 19:50 UT, 5:50 in the morning in New South Wales, "when the visitors centre is closed and unstaffed."

"We conclude the evidence in favour of FRB 010724 being a genuine FRB is strong," they wrote, and in their summary: "The peryton-causing ovens on Parkes site could not have produced FRB 010724."

The perytons had other tells. They came in office hours, mostly on weekdays, peaking around lunch. They did not follow the exact curve that cold plasma between the stars draws. And since 2015 fast radio bursts have been caught by arrays of dishes, and pinned to distant galaxies, which no oven could do.

What the oven does not explain

The ovens explain where the perytons came from. They do not explain the perytons.

Petroff's team found the source and said plainly what they could not account for. "The physical process that generates the swept or 'chirped' emission that defines these perytons is obscure," they wrote. "The duration of the perytons is also a puzzle." The oven's supply should die away within a few 20 millisecond mains cycles, the team noted, yet each peryton took about a quarter of a second to cross the band.

Then there is the cluster of 1998, sixteen perytons inside seven minutes on 23 June, the eleven of Burke-Spolaor's and the five Kocz added. Their spacing was 22 seconds, give or take a third of a second. The team's reading was an oven run below full power, cycling its magnetron on and off, and they conjectured something in the door, perhaps foil, letting the radio out without anyone opening it. "We have yet to devise an acceptable test of this scenario," they wrote.

And the number that worried everyone could not clear the kitchen. The paper says the burst's 375 is "entirely consistent" with the Woolshed perytons near 400. The sweep ruled out one oven, the one in the visitors centre. The beams, the pointing and the hour did the rest.

They did not find the physics of the chirp.

Which leaves the oldest and strangest fact in the case standing exactly where it was: a household appliance, opened early, draws a curve that astronomers had taken for the mark of the universe itself.

The kitchen and the sky

The suspicion is as old as the field. The Cambridge team that found the first pulsar in 1967 wrote in Nature that its signals had at first suggested man-made transmissions, and ruled those out before they published.

The answer gave Parkes a test. Standard practice at the observatory was already to keep the ovens off when observing in the 2.4 gigahertz band; now the 2.3 to 2.5 gigahertz monitor gave a test the observatory could run, and Petroff's paper proposed it: a peryton comes with emission there, a fast radio burst does not. At Arecibo, Robert Kerr told National Geographic in 2015, "Microwave ovens are a problem for us - and none exist on site. They are prohibited."

Petroff took issue with the coverage. "From the media's perspective this story was apparently too tempting not to spin," she wrote. "Alas we were never looking for extra-terrestrial life, studying alien signals or confusing astronomy with gastronomy." In 2023 she told Popular Science: "I don't think I've ever had as satisfying of a moment in my research."

The ovens were more than 27 years old in 2015. The Woolshed oven's magnetron was a type 2M210-M1. Their chirp has a name, a date, a building and a mug of water.

It does not have an explanation.

The answer to the question in the title is yes, for seventeen years, and no, for the one burst that mattered most.

Questions This Leaves You With

Is there anything a person actually does with this?

Not much in a daily sense, and it is fair to say so. Your microwave is not signalling anyone. What the story does offer is a habit worth borrowing: when something strange keeps happening, write down when. The perytons gave themselves away by keeping office hours and peaking at lunch, years before anyone tested a kitchen. The larger question, how a civilization this noisy will ever hear anyone else, is not yours to answer alone.

Does this sit against a religious account of the world?

It cuts both ways. Some will read it as a lesson in humility: the wonder people saw in the sky was, this time, their own lunch. Others will read it the other way: the care that cleared the Lorimer burst is what lets the real flashes stand, and those are stranger than anything in the kitchen. Neither reading needs the oven to mean more than it does. This site will not tell you which reading to hold. That is yours.

Does it change what a human life is?

Put yourself at the bench in March 2015, holding an oven door, waiting to see whether a seventeen-year question ends when you open it. One reaction is that it makes the work feel small: the mystery was a kitchen. The other is that it makes it large: a handful of people, by patience and a well-placed monitor, cleared the ground so that the founding event of a field could stand. Which did you feel first?

What would change your mind, in either direction?

Name it now. Toward the oven explaining everything about perytons: a lab measurement of a magnetron shutting down that draws the same chirp and lasts the same quarter second. Toward something left over: a peryton with no 2.4 gigahertz emission beside it, or one at night with every oven off. And for the bursts themselves: a fast radio burst caught by one dish alone, with no second telescope, would again have to answer the question this story asked.

Where does the open problem actually sit?

Not in where the perytons came from. That was settled in March 2015, with the door, the mug and the monitor. It sits in how. The paper that caught the machine in the act called the process behind its chirp obscure and its length a puzzle. You are looking at a solved case with the physics still missing.

The kitchen showed how a convincing imitation could be made on Earth. The geometry showed how far that imitation could reach, and where it stopped.

So the next time a curve appears on a single dish, what would you ask for first: a second telescope, a log of every machine on site, or the monitor's trace at 2.4 gigahertz? Then go back to the Lorimer burst and see whether that first curve looks different now that you have seen how one can be made on Earth.