US2138920A - Secondary emission tube and circuit - Google Patents

Description

Dec. 6., 1938. H. E. HOLLMANN SECONDARY EMISSION TUBE AND CIRCUIT Filed Feb 17, 1937 INVENTOR HANS ERICH HOLLMANN A ORNEY Patented Dec. 6, 1938' 2,138,920 SECONDARY EMISSION TUBE AND= CIRCUIT Hans Erich Hollmann, Berlin, Germany, assignor to Telefunken Gesellsohaft fiir Drahtlose Telegraphic, in. b. H., Berlin, Germany, a corporation of Germany Application February 17, 1937, Serial No. 126,111- In Germany April 23, 1936 3 Claims.

Ordinarily the electron tubes serving for the production of electrical oscillations operate with glow emission, i. e., the electrons sustaining and controlling the oscillation mechanism are emitted by a glow cathode whose heating in case of larger tubes, requires considerable power. recent times, tubes have become known which operate exclusively with secondary electrons instead of glow cathodes, said secondary electrons are released from the atomic structure of the cathode surface when primary electrons impinge thereon. In these tubes the cathode or cathodes remain cold so that no special heating is required.

In order that an electron current be continuously and exclusively sustained by secondary emission, it is necessary that all or part of the electrons once produced, again and again release new secondary electrons, and this in greater numbers or at least in a number equal to that being lost by the transit to the anode or to other absorption electrodes. In order to providesuch continuous production of secondary electrons, the arrangement schematically shown in Fig. 1 has been developed. Herein, items K1 and K2 are two fiat cathodes which, however, do not emit as such but are covered with layers having secondary emission property of great yield such as cesium oxide layers. As is known, such layers are capable of giving up a number of secondary electrons, that is, ten to twenty times the number of primary electrons impinging thereon with a sufficiently high velocity. Between the two cathodes K1 and K2, a suitable anode A is arranged which must be wholly or partially permeable as regards fr the electrons passing towards it. In the example herein shown, a grid-shaped anode A is shown catching a small part of the electrons, while the greater part passes unhindered through its meshes. In principle, the scheme is not changed if the real grid anode is replaced by a virtual acceleration grid, for instance in the form of an anode ring or the like surrounding the discharge space. In the case of a flat electrode structure, it was found suitable to concentrate the electron current by means by an additional magnetic field whose lines of force extend in the direction of the electrical field lines, i. e., from cathode to cathode. However, by means of special electrode arrangements, it is also possible to obtain the 5D- desired concentration by purely electrostatic means.

The performance of the automatic secondary emission and excitation of oscillations is as follows: When applying the plate potential from 55 one of the two cathodes a few electrons, for in- Lil However, in

stance photo-electrons, may at first be emitted. These electrons are being accelerated by the grid, and a part thereof arrives at the opposite cathode, where they may release, for instance, ten times as many secondary electrons. These return again to the first cathode and produce thereat a still greater number and so on. In this way the total emission is gradually enhanced until a final stationary state is reached, determined either by the limited yield of the cathodes or else by the space charges.

The excitation of oscillations in such an arrangement can be conceived in the simplest manner as pendulum movements of the electrons, as known from the Barkhausen retarding field method, more especially from the push-pull retarding field tubes. The only difference existing resides in that the electrons do not reverse their course in front of the cathodes on both sides, and that therefore a certain electron cloud can. oscillate b-ack and forth several times, but instead thereof, in each half cycle, a new electron cloud is produced and introduced in the oscillation performance. There is connected therewith a synchronous control of the emission. in the rhythm of the travel times of the electrons thereby obtaining an especially favorable excitation.

In the principal form so far described, the mentioned building up performance could, however, not yet be obtained, since the secondary electrons emitted by K1, for instance, and after having passed between K1 and A, would be decelerated in the equally intensive but oppositely directed retarding field between A and to such an extent that they arrive at K2 with zero velocity. However, in this case a release of secondary electrons is not possible.

In order that a certain velocity is retained in the electrons when impinging on K2, the latter must be positively biased to a slight degree. In the following half cycle, however, K1 on the contrary must have a positive bias potential so that the same conditions prevails also for the secondary electrons passing from K2 to K1. In the hitherto known arrangements, this operating condition is fulfilled in that an alternating potential is applied between the two cathodes, and whose alternation conforms to the travel period of the electrons from one cathode to another one. This auxiliary potential can be supplied by a separate control transmitter. A simpler means resides, however,- in that between the two cathodes a suitable oscillatory circuit LC, according to Fig. 1, or a Lecher system, or the like, is inserted. The latter'is then caused to oscillate by the periodic electron current, so that the transmitter produces itself on the basis of a feed-back like performance the control potentials necessary for its operation. Only under this presupposed condition is the described building up of the emission possible, and plate current and excitation of oscillations become a maximum.

Now in the present invention a new way is shown in order to retain in the electrons at the transition from one cathode to another one, the kinetic energy necessary for the release of secondary electrons, whereby various advantages are obtained over the hitherto known arrangements of the type of Fig. 1. Since, in fact, in the acceleration and the subsequent deceleration of the electrons it is not the absolute electrode potential which plays the important role, but the potential differences, or to be more exact, the field intensities existing in the two electrode spaces, the smaller deceleration during the second part of each alternation of the pendulum movements of the electrons is effected in accordance with the invention, in that instead of the cathodes the anode receives an alternating potential controlling the discharge. In analogy to the preceding observations, the control potential must in this case have the following course: As long as the electrons are, for instance, between K1 and A, i. e., in the first half of the half of the pendulum movement, the plate potential may have a certain definite value. As soon as, however, the electrons have passed the grid, the plate potential must be lowered so as to reduce the intensity of theretarding field to such a degree that the electrons impinge on K2 with the necessary residual velocity. In the subsequent quarter cycle of the pendulum movement of the electrons during which the newly released secondary electrons fly towards A in the opposite direction, the grid potential can again resume its original value but it is then necessary that in the last quarter cycle, i. e., when the electron pass through the path AK1, the grid potential is again reduced in order that the said excess velocity also of the electrons passing towards K1 be retained- A fundamental difference as compared with the control potential between the cathodes, it is thus seen that the control potential of the grid must vary during each half cycle, i. e., during each electron movement from one cathode to another one, from a maximum to a minimum and again back to a maximum which is at fully 360, or in other words, the control potential at the grid must have twice the frequency of that which would be used if the control potential were applied to the cathodes.

To further explain this performance it is referred to the diagram shown in Fig. 2. Herein,

the lower part shows the movement as to time of two secondary electron clouds in the direction from K1 to K2 (full line) and in the direction from K2 to K1 (dash lines). At the place above the tube there is shown the apertaining course as to time of the grid potential. In the first quarter cycle of the electron oscillation considered from K1 back towards K1, the grid potential is negative, and the electrons are accelerated in a corresponding degree. During the subsequent quarter cycle, the alternating grid potential will be reversed, so that the electrons receive in the average a weaker deceleration than would correspond to the preceding acceleration, and the electrons impinge on K2 with the velocity component remaining as difference. For the new secondary electrons now released-by K2 the same is true accordingly, but now the phase of the alternating grid potential is reversed in accordance with the reversed direction of the electron movement.

At this place it may be remarked in principle that the described performances are not necessarily confined to the double frequency of the control potential of the gird, but can. also be applied in an analogous manner to higher harmonies of the pendulum-movements of the electrons as long as the fundamental principle is retained, namely that the electrons after passing through the grid-anode are less retarded in both directions than they have been accelerated previ ously. In practice, the described performance is of importance in various respects. In the first place,'it is obvious that if the transmitter produces itself in a corresponding manner its control potential, the resonance system placed at the grid can be tuned to the double frequency which the resonance circuit, situated between the cathodes, would have. The new arrangement thus furnishes principally a frequency which is twice that produced with the same tube dimensions and operating conditions in the hitherto customary arrangements. Furthermore, the new arrangement owing to the effect described, is well suited for frequency multiplication. To this end, the pendulum movements of the electrons are controlled in the hitherto customary manner by an alternating potential between the cathodes, whereafter in accordance with the invention, in the grid circuit a resonance circuit tuned to a multiple of this control frequency can be excited. Instead of taking the control frequency from a special control transmitter, it can also be produced'in the cathode circuit by the self-excitation above described. This method has a particular advantage when ultra-high frequency oscillations are to be stabilized by a quartz or turmaline crystal. The control crystal may then, of course, be inserted in the cathode circuit and stabilizes at first the pendulum movements of the electrons, whereafter on the basis of the multiplication effect afore described in the grid circuit, an oscillation can be derived whose frequency is twice the crystal frequency or a multiple thereof.

As mentioned above, Fig. 1 illustrates a known oscillator arrangement, while Fig. 2 is given to explain the time relations of the electrons in their travel between the cathodes with respect to the potentials applied to the electrodes. A circuit arrangement for-the production and multi-- plication of high frequency oscillations, in accordance with the principles of the invention, is shown in Fig. 3.

Fig. 3 shows my new arrangement wherein the grid electrode itself is designed as the resonance system to be excited. To this end, the grid leadin is connected to the center of the grid A in which an oscillation node exists, while at the grid ends oscillation loops appear in accordance with the voltage distribution shown in dash lines. Principally, the entire tube may obviously also be constructed with cylindrical electrodes whereby the grid anode consists of an open or closed helix. The ratios between the radii of the three electrodes are to be so chosen preferably that the electron travel times are possibly the same in the space inside and outside the grid. Otherwise all features known from the ultra short wave field can be applied to the new arrangement.

Regarding the new application of the secondary emission tube, it is likewise the case that the real grid shown for the sake of simplicity in the present example can also be replaced by any virtual grid without departing from the principle of the invention.

What is claimed is:

1. In combination, an electron discharge device comprising an evacuated envelope containing a pair of parallel surfaces capable of emitting electrons on impact, said surfaces being oppositely disposed and parallel with respect to an intermediate grid collecting element, said grid having an overall length equal to half the length of the operating wave, and means for applying to the center of said grid a potential which is positive With respect to said surfaces.

2. In combination, an electron discharge device comprising an evacuated envelope containing a pair of parallel surfaces capable of emitting electrons on impact, said surfaces being oppositely disposed and parallel with respect to an intermediate grid collecting element, said grid having an overall length equal to half the length of the operating wave, a direct connection between said surfaces, a circuit coupled between the center of said grid and said surfaces including means for applying to the center of said grid a potential which is positive with respect to said surfaces.

3. In combination, an electron discharge device comprising an evacuated envelope containing a pair of parallel surfaces capable of emitting electrons on impact, said surfaces being oppositely disposed and parallel with respect to an intermediate grid collecting element, said grid having an overall length equal to an odd multiple including unity of half the length of the operating wave, and a circuit coupled between an oscillation nodal point on said grid and said surfaces for applying to said grid a potential which is positive with respect to said surfaces.

HANS ERICI-I HOLLMANN.

Images