Access to Energy

HOW A MAGNETRON WORKS

The figure shows a cross-section through a magnetron with the resonant cavities linked to a central cavity by narrow channels. The material was a copper alloy and the cylinder was about ¾" high, with flat ends that fitted into the gap of a big, heavy and powerful permanent magnet close to a horse-shoe type.

GRAPHIC: Magnetron cross-section showing resonant

cavity, grounded plate, electrons, cathode at -14,000 V potential and cooling fins

Of course, the resonant cavities are not enough, you have to ex-cite them, build up the resonant wave and keep it going. For this purpose, first of all there was a cathode 14,000 V below the grounded main block. It was a sign of the times that they let 18-year olds play with a "hard" source (one not diminishing in voltage when under load) of 14 kV. One Canadian boy somehow got into contact with it, yelled out, and one who was watching him claims that he saw a blue corona round the middle of his body. Everybody rushed around him asking "Are you OK?" He answered, "I am fine, thank you," took a step forward, and collapsed unconscious on the floor. He was taken to Station Sick Quarters, where he came to, and after a day or two of flirting with the nurses he walked into the radar workshop as good as new. Somehow the current through him must have missed his heart, or he would have been electrocuted.

Anyway, the cathode in a vacuum tube is heated red-hot so that it emits electrons, and if it were not for the magnet, they would go by the shortest distance, a short straight segment, to the positive plate. (Zero volts is 14 kV positive with respect to -14 kV.) But the magnetic field perpendicular to the page (not shown) curves their path so that it turns into a spiral arc before it reaches the plate.

But there is something else. As the cavities oscillate, the ends of the channels go positive and negative as shown. (How they first started oscillating is a chicken-egg question, for which I have no space.) Perhaps you object that the copper would short out this plus and minus, but at frequencies of some 1,000 MHz the flow of current is impeded by more than just ohmic resistance¾again for reasons for which I lack space.

Now as the electron stream goes past these pluses and minuses, the electrons near a plus are attracted and accelerated, while those near a minus are repelled and slowed down. The electrons in the stream therefore no longer form a regular (actually random) procession, but bunch up into "electron bunches," and it is these electron bunches that feed energy into the cavities and keep them oscillating by reinforcing the plus and minus as they fly by in synchronized clouds.

The output is taken from one of the cavities by a probe (not shown) in the form of a wire hook, which feeds the high frequency through a wave guide to the application, here to a radar antenna.

Last time [AtE Sep 92] we talked about feedback. All oscillators work by taking the output, amplifying (enlarging) it and feeding it back to the input in phase with it, thus reinforcing it. (This is called "positive" feedback in distinction to negative feedback that is designed to prevent oscillations such as in Watt's governor dis-cussed last time.)

In the case of a magnetron, the "regulated system" are the resonant cavities, the input are the electromagnetic oscillations in-side it, and both amplification and feedback is provided by the electron bunches.



 • Not by the Free Market Alone
 • HOW THE ATLANTIC WAS WON
 • HOW A MAGNETRON WORKS
 • NEW APPLICATIONS OF MAGNETRONS
 • MEET SEPP
 • ECHOES AND UPDATES
 • STARK RAVING MAD
 • GOOD READING
Vol. 20, No. 2

Newsletter: Access to Energy Newsletter Archive
Volume: Volume 20
Issue/No.: Vol. 20, No. 2

Date: October 01, 1992 10:48 AM
Title: Not by the Free Market Alone

Copyright © 2004 - Access to Energy Newsletter Archive
All rights reserved.