Some Electron Multiplier Tubes

Author: J.B. Hoag

The fact that the number of secondary electrons knocked out of a metal plate can exceed the number of incident or primary electrons forms the basis of multiplier tubes. In Fig. 15 G, electrons from the cathode K are controlled by grid C'G and accelerated by the second grid AG. They are directed by electrostatic fields onto the positively charged metal plate 1, where they give rise to secondary electrons. In the figure two secondary electrons are ejected for each primary electron. The secondaries are attracted to the still more positive plate 2, from which four secondaries are ejected. The latter pass to the plate P and to the output. In this case, the current (number of electrons per second) is multiplied fourfold.

Fig. 15 G. Principle of a multiplier tube

A small voltage on the first grid of Fig. 15 G is effective in controlling the number of electrons which pass through it to the next electrode. The effectiveness is tremendously augmented by the fact that the electrons which do pass through the control grid are multiplied many fold. In other words, a very small voltage on the control grid will cause a very large change in the output plate current. The mutual conductance (g_m = change of plate current per volt change on the grid) of these tubes can be varied greatly by changing the voltage applied to each multiplying stage because this changes the number of secondaries per primary.

A more practical form of a multiplier circuit is shown in Fig. 15 H, where a single high-voltage supply source (3,000 volts) is used.

Fig. 15 H. A more practical form of multiplier tube

The proper voltages on each multiplier stage are obtained from the potential drop in the series of resistors RR.

The absence of coupling stages between one tube and the next makes possible the amplification of a succession of pulses which are very close together, or of very high frequency (microwaves). At the very high frequencies, the time for the electrons to travel from one electrode to another becomes an important fraction of the period of the oscillations. This results in a loss in amplification. The maximum theoretical frequency with present-day tubes is about 2 · 10⁹ cycles per second.

In the orbital-beam multiplier tube of Fig. 15 I, the electrons are deflected in a circular path by a positively charged focusing electrode. The tube should prove useful for frequencies of the order of 500 MHz (wave-length = 0.6m).

Fig. 15 I. An orbital beam multiplier tube

Further discussion of multiplier tubes will be found in the chapter on photoelectric cells.