I'm curious why magnetrons from a home microwave like in the image below can have an aperture between the magnetron cavity and the waveguide that is significantly smaller (2-6mm usually) than the wavelength of the 2.45GHz radiation (122.45mm roughly.) Is this just the point at which the wave can leave the cavity without being obstructed based on the geometry, are the uncollimated waves coming out of it at a relatively steep angle at that point, something else, etc?
Reading Wikipedia I get the next information: "These devices operate ..., using the ballistic motion of electrons in a vacuum under the influence of controlling electric or magnetic fields ... These devices work in the density modulated mode, rather than the current modulated mode."
This means that they work on the basis of clumps of electrons flying ballistically through them, rather than using a continuous stream of electrons." This means that microwaves are modulated radiation and the periodicity of the electric and magnetic field components of this EM radiation is induced by bunches of electrons.
This induction process is in principle the same process as in an antenna rod, where the antenna generator push and pull electrons forward and backward. Accelerated electrons emit EM radiation. In an antenna rod the electrons get accelerated linear, in a microwave generator the electrons get accelerated by circular motion. (And the periodicity of such a EM field let the water molecules oscillate.)
The point is that the electrons emit photons -the units of the microwaves - in different from the generator frequencies, such as X-rays. Understanding this it is easy to understand why a much smaller slit or hole as the calculated wavelength is enough to let through microwaves. Simply longer wavelengths than infrared and perhaps terahertz waves are not used for commercial use. What I'm curious is, what is the longest measured wavelength for excited molecules.