Modern microwave ovens do an excellent job of containing microwaves, with leakage levels well below established safety standards. However, while the miniscule amount of leakage has no impact on human health, it has the potential to interfere with the growing multitude of wireless devices operating in the same environment.
Cell phones, Bluetooth devices, Wi-Fi devices and other wireless electronics that operate at similar frequencies are all vulnerable. Interruptions among the devices are inevitable, says Ron Gilgenbach, professor of nuclear engineering and radiological sciences at the University of Michigan. That’s because a broad range of electronics operate in the unlicensed electromagnetic ISM band of 2.4 GHz to 2.45 GHz — the same frequency range utilized by magnetrons in microwave ovens.
But the regulations that make that band a free-for-all could someday change, and the manufacturers of the electronics equipment that operate in this frequency range must take into consideration how they will react if the FCC decides to step in and regulate it, placing limits on microwave emissions.
The Bluetooth and IEEE 802.11 electronics operating in this frequency range provide for spread spectrum communication, allowing the signal to jump about to different frequencies within the spectrum. So if there is a channel experiencing interference from a microwave oven, the device can communicate by jumping to another channel.

Mystery of EM noise

The electromagnetic noise emitted by microwave ovens, however, is a subject that is not as easily characterized.
“The noise mechanism for magnetrons is not well understood,” Gilgenbach says. “Since World War II, researchers have been trying to find ways to reduce the noise of the magnetron. I, personally have been studying it for more than a decade, and I still don’t totally understand it.”
The magnetron is one of two types of electron beam devices. In the common magnetron configuration, the magnetic fields are crossed, or perpendicular to one another. In other types of devices, such as klystrons, the electric and magnetic fields are parallel to one another. The perpendicular configuration emits more electromagnetic noise than its parallel counterpart, and its orbits are more complicated. More than 50 years after the invention of the magnetron for use in military radar, its physics remain a mystery, Gilgenbach says.
With funding from the U.S. Air Force Office of Scientific Research, Gilgenbach, U-M Professor of Nuclear Engineering and Radiological Sciences Yue-Ying Lau and graduate student V. Bogdan Neculaes, went about the task of determining how to reduce the noise level of magnetrons used in radar systems.
With the rising threat of terrorism, the office commissioned the work in an effort to develop a powerful, yet low-noise radar system, that would enable the U.S. military to obtain a higher resolution, and consequently track smaller objects.
For purposes of their experimentation, the research group used microwave oven magnetrons because the physics of the magnetrons are the same as those employed by radar systems. The researchers also knew that scientific research on an actual radar system would more than likely be stamped classified, keeping the findings out of the public domain.
The group’s findings were published in a series of papers and two patents. The first, paper “Low-Noise Microwave Magnetrons by Azimuthally Varying Axial Magnetic field,” described their discovery that by introducing magnetic modulations to the system, they could immediately set the magnetron at the correct frequency, decreasing emission of electromagnetic noise.

Emphasizing imperfection

Fig. 2. Simulation of electron dynamics in a six-cavity magnetron (electron spokes shown in red).

Previous attempts to decrease the noise of the magnetron were similar in two ways — the experiments involved trying to make the magnetic field more perfect, and — they did little to affect the noise level of the device. When the U-M research group went about the task, they decided to try something different.
“For 50 or 60 years everyone has been trying to make the magnetic field more perfect in these devices, and we found that by making the magnetic field less perfect, it actually improved the noise performance,” Gilgenbach says.
Inside the microwave magnetron are electrons that spin in a circle about a cathode, and on the outside of the cathode is an anode. The anode is comprised of cavities that exhibit a resonant microwave frequency, which interact with the spinning electron beam. Instead of spinning around as a uniform beam, the electron beam, spins in spokes, much like the spokes on a wheel. The spokes spin around and slosh in and out of the cavities, creating microwaves.
The researchers introduced magnets to the magnetron to alter the modulations of the magnetic field, and were surprised with their results.
“We found that by making the magnetic field have modulations as the electrons spin around, we could actually throw the electrons into the spokes right away. Normally it takes the electrons longer to form these spokes, and during that time, they are generating microwave noise and emitting the wrong frequencies,” Gilgenbach says. “We throw the electrons into the spokes instantaneously with the magnetic field. Consequently, they are immediately emitting the correct frequency.”
The simple magnetic field modification was shown to significantly reduce the microwave noise in magnetrons by azimuthally varying the axial magnetic field. Without the modulated magnetic fields, the random motion of electrons generates random, noisy microwave frequencies. The modulated magnetic field ensures that the desired microwave frequency is “primed” to start immediately before the unwanted frequencies can grow. This noise reduction technique is effective in fresh or aged magnetrons and cleans the spectrum especially well during the start-oscillation condition.
By modulating the magnetic field, the researchers were able to allow the system to start up much more rapidly at the proper frequency, eliminating noise by 30 dB, even when close to the carrier frequency. The physical mechanism of the electromagnetic noise reduction is not completely understood; one candidate is the perturbing effect on recirculating electrons in the magnetron.

Magic combination

With 4 perturbing magnets, the signal is noise-free.

In their second round of experimentation, the researchers went one step further, determining the optimal number of magnetic variations. Their findings, published in a second paper, “Simulation of Rapid Startup in Microwave Magnetrons with Azimuthally Varying Axial Magnetic Fields,” indicates that the optimal number of magnetic variations is equal to the number of electron spokes that are required to create the proper frequency.
U-M has patents on both technologies and is seeking licensees in the consumer market. The university is especially interested in licensing the technologies to the manufacturers of microwave ovens.
“We’re looking for a licensee, and we’re very interested in an industrial partner to continue this research to apply these scientific results to actual microwave ovens,” Gilgenbach says. “We need to find ways to reduce electromagnetic noise in microwave ovens in case the FCC decides to place limits on the electromagnetic emissions. I think companies need to be ready to address that situation before it happens.”