US2623129A - Thermionic tube for amplification of ultrashort electric waves - Google Patents

Description

A. LERBS 2,623,129 xomc TUB FOR AMPLIFICATION ULTRASHORT ELECTRIC WAVES THERM Dec. 23, 1952 3 Sheets-Sheet 1 Filed May 25, 1949 like". 23, 1952 A. LERBS 2,623,129 THERMIO c TUBE FOR AM FICATION H 0F UL SHORT ELECT WAVES I Filed May 25, 19,49 3 Sheets-Sheet 2 INVENTvAZ .AHFRE) 512.25 2 5 my fla s/v75 5 Sheets-Sheet 5 A. LERBS THERMIONIC TUBE FOR AMPLIFICATION OF ULTRASHORT ELECTRIC WAVES "Dec. 23, 1952 Filed May 25, 1949 I/v VISA/70?: 94 FRED 1.553s

of the acceleration voltages.

Patented Dec. 23, 1952 THERMIONIC TUBE FOR AMPLIFICATIQN OF ULTRASHQBT ELECTRIC VJAVES Alfred Lerbs, Paris, France, assignor to Compagnie Generale de Telegraphie SansFil, a corporation of France Application May 25, 1949, Serial No. 95,188

' In France June 12, 1948 1 Claim. 1,

My invention relates to anew type of: tubes called progressive-wave tubes which, although they are more particularly used as amplifiers, can also be used for producing and receiving ultra-short electric waves.

Ifhe heretofore known tubes of this type comprise an input terminal E and an output terminal S between which extends a delay line through which a progressive-wave travels at a phase velocity u which is parallel to the delay line and the direction of which is the same as that of the phase of the progressive wave, is located throughout its travel from E to S in the high-frequency field produced by the progressive wave. For the operation of such tubes, it is necessary for the velocity in of the electrons to be equal to the phase velocity Up.

This condition (09:01:) is a drawback in the heretofore known types of tubes, since the velocity of the electrons is proportional to the square However, a decrease in the, voltages is only permissible in the case in which the phase velocity up is also decreased, which cannot be effected at will. If, for

example, the delay line is formed by a wire wound in the shape of a cylindrical helix H, in order to. decrease the phase velocity u it would be necessary to reduce the distance a between the convolutions, or increase their diameter d (see Fig. 1 in which the arrow shows the direction of the electron beam and of the phase of the progressive wave, Q is the cathode, E and S are respectively the input and the output terminals of the delay line). But these two means produce a decrease in the necessary energy exchange between the electron beam and the progressive wave, because a decrease in the distance a causes the field of the progressive wave to shift in the direction of the contraction of the helix and an increase in the diameter d increases the distance between the axis of the elec' tron beam and the delay line. Thisdrawback is observed not only in progressiverwave tubes with a, straight electron beam (shown schematically in Fig. l), but also in progressive-wave tubes with a magnetic field, in which a circular movement is superimposed on the rectilinear movement, this latter movement talring place in the direction of the phase of the progressive wave.

My invention enables the said drawbaclrs to be obviated, since therein the voltage of the anode is independent of the phase veiocity. No o n ne h r e be taken of t at e oc ty in choosing the distance between the convolu- The electron beam, the path of tions, so that the exchange of energy between the field of the progressivewave and the electrons can be considerably increased. Furthermore, it procures the advantage of not having to use special electron optical devices and of enabling without difficulty considerable currents to be made to flow through the tube according to the invention.

My invention has for its object a tube which procures such advantages and which comprises an emitting cathode and an anode constructed in the shape of a delay line provided with an input terminal and an output terminal, a wave being adapted to travel progressively through said line, and an electric field which extends between the, cathode and the anode and is directed towards the anode. According to the invention, the emitting cathode and the anode are arranged coaxially so that the axis of the cathode extends in the direction of propagation of the wave in the anode, said anode and said cathode both being preferably of linear shape.

The essential difference between the new tube and those heretofore used consists in the fact that there need not be a movement of electrons in the direction oi the phase of the progressive wave, but that there is a periodic reciprocating movement of the electrons in a direction at right angles to that of said phase. The exchange of energy between the electrons and the progressive wave is effected by means of the transverse component of the high-freduencyfield. Fig. 2 shows the instantaneous distribution of the highfrequency field produced by the progressive wave of said transverse component as it propagates along the cylindrical helix H. The line a- -a1 indicates the position of the longitudinal component and the direction of the movement of the electrons in the tubes heretofore used. In the arrangement proposed herein, use is made of the transverse component of the field, said transverse component being at right angles to said line and producing a periodic reciprocating movement of the electrons parallel to b-b1. The necessary energy for the movement of the electrons is supplied by a source of direct current. If suitable means are chosen so that the largest possible number of electrons can transfer the energy obtained from the source of direct current to the high frequency field and that on the. other hand a mall number o e ec receive highfrequency energy, an amplification of the progressive wave takes place.

The invention will be more clearly understood by means of the accompanyingdrawingswhich show, by way of nonlimitative examples, several embodiments thereof.

Fig. 1 shows a diagrammatic arrangement of th known progressive-wave tube.

Fig. 2 shows the distribution of the field.

Figs. 3 and 31) show the diagrammatic arrangement and Fig. 3a the practical construction of the tube according to the invention.

Fig. 4 shows the distribution of the field in such a tube.

Fig. 5 is intended to enable a known magnetron to be compared with Fig. 32).

Fig. 6 shows a distribution of the field.

Figs. '7, 8, 9, 10, 11 and 12 show diagrammatically modifications of electrode and supply arrangements.

A tube according to the invention is shown diagrammatically in longitudinal section and in transverse section in Fig. 3 and in a more complete manner in Fig. 3a. E and S denote the input terminal and the output terminal of the delay line which, in this embodiment, is of the shape of a cylindrical helix. On the axis of said helix is located the cathode C heated by the auxiliary battery Ch. The invention also provides, according to a modification, a constant magnetic field B, the direction of which is shown by the arrow B and is parallel to the axis. With respect to the cathode C, the helix H is at a positive potential Va and therefore acts as an anode. So long as B=0, the electrons travel directly towards the anode. If, on the contrary, B is sufliciently great, the electrons travel round the cathode and furthermore effect a periodic reciprocating movement between H and C, see Fig. 3; the transverse section in the same figure shows the trajectories I, 2, 3 of the electrons. During this movement, the electrons remain in the same section, i. e. they do not move along the axis of the cathode or of the magnetic field. Fig. 3a shows a more complete view of such a tube, wherein the same reference numerals denote the same elements. The input E and the output S of the delay line are effected by means of coaxial cables L1 and 1.2 through the vacuum-tight bulb V of the tube. B1 and B2 show schematically the pole-pieces of an electromagnet adapted to produce the magnetic field B parallel to the cathode. U1 and U2 denote the leading-in wires of the cathode C which are intended to be connected to the terminals of a cathode heating battery, the helix H and the cathode C being connected to a source Vd of direct current voltage according to Fig. 3 and not shown in Fig. 3a.

Fig. 4 shows the instantaneous distribution of the high-frequency field of such a tube. The lower curve shows the corersponding instantaneous values of the transverse component along the helix. In reality, this diagram moves from E towards S, whereas the movement of the electrons takes place in the same transverse section in which they were emitted by the cathode. The transverse component of the high-frequency field, which acts on the electrons, therefore varies in cadence with the high-frequency. Consequently, during their pendular movement between H and C, the electrons are subjected to almost the same conditions as in a magnetron with a non-slotted anode, which is shown diagrammatically in Fig. 5. A is the anode, in the axis of which the cathode C is located; B denotes the magnetic field; I, 2, 3 are the trajectories of the electrons. The oscillating circuit represented by the capacitors C1, C2 and the inductance L is located between C and A; in the. radial direction.

between C and A, is located a stationary highfrequency field. The difference between the device according to Fig. 3 and the one shown in Fig. 5 essentially consists in the fact that the latter is capable of self-oscillation owing to the reaction produced by the stationary oscillation, whereas in the device of Fig. 3, at any rate in the case in which reaction between E and S is prevented, it is necessary for an alternating current power to be applied at the input E and to be propagated as a progressive wave towards S, its being possible to use this device to obtain an amplification. However. the electronic mechanism of both these devices is the same in principle and a detailed explanation is not necessary. The same applies to the electronic mechanism of the examples given hereinafter.

In a similar manner, it is necessary in the case that corresponds to Fig. 3, for the duration of the pendular movement to be equal to the periodicity of the alternating current voltage.

In the formula K A cm. (1)

A is the length of the progressive wave in centimetres, measured in air, K is a coefilcient of about 12,000, which may vary within certain limits according to the distribution of the field in the discharge space. In the first place the magnetic field is therefore determined by the wavelength, while the anode voltage U... which exists between H and C can be approximately calculated if the value of B and the radii of the transverse sections of H and of C are known:

In order to obtain an efiicient electronic mechanism, it is necessary for the electrons to be moved away from the discharge space after they have efiected a number of oscillations, and this can be obtained by imparting to the electrons an additional movement in the direction of the anode. For this purpose, means can be used which are known in connection with magnetrons provided with a non-slotted anode, viz. a slight inclination of the magnetic field with respect to the cathode or the use of a slightly distorted magnetic field. It is also possible to use a distortion of the D. C. electric field, such distortion being effected either by means of special additional electrodes, or of a geometrical asymmetry in the mounting of the tube.

Although the use of a delay line is not necessary a priori, it is nevertheless necessary to have recourse thereto for practical reasons. It might be suificient, in principle, to cause the progressive wave to travel through a coaxial line, the inner conductor of which would act as the oathode and the outer conductor would form the anode. However, in order to obtain a sufficient damping of such a line and make it sufficiently aperiodic. or in order to provide therein a number of wavelengths equal to that of a delay line, its length would have to be greater than that of the delay line in a ratio equal to the ratio between the velocity of light and the phase velocity. which in practice means from 10 to 20 times greater. But this is not permissible for practical reasons, in consideration of the cathode for example.

If the distance a between the convolutions of the helix H is great as compared to the distance CH, the distribution of the D. C. electric field becomes non-uniform in the direction of the axis,

Guuss which may possibly make the electronic mechanism less eflicient. Fig. 6 shows such a distribution of the field. The dotted lines represent the lines of force. However, it is possible to compensate completely the effect of the distance between the convolutions by surrounding the helix H with a cylinder P connected to a higher voltage than that of the anode. If the voltage is so chosen that the helix can be considered as an equipotential electrode, the distribution of the field remains constant in the direction of the axis and corresponds to Fig. 7. By choosing other values for the voltage, fields can be obtained of an intermediate shape between the shapes shown in Figs. 6 and 7. The electrode P can be used to act on the electronic mechanism and to adjust same.

The cathode C and the outer cylinder P may also be permuted as shown in Fig. 8, so that use is made of the portion of the high-frequency field located outside the helix. Such an ar rangement should be considered in particular if there is a considerable dissipation of energy in the tube and if an inadmissible heating of the cathode due to the bombardment of the electrons (retroactive heating) occurs. By means of such an arrangement, the possibility can be readily obtained of making only a portion of the cathode operative for effecting the emission, for

example a narrow strip C1 thereof (Fig. 9) parallel to the axis. Although the electrons remain in the same transverse section, they move within said section, about the axis in the direction of the arrow 2, which may possibly cause the electrons to rebound on the non-emitting portion of the cathode.

The progressive-wave tube with a magnetic field hereinbefore described is only a non-limitative example of the embodiment of the idea of the invention. It is also possible to consider the use of electronic mechanisms in which a magnetic field is not necessary for obtaining an oscillatory movement of the electrons. Such a pendular movement can be obtained, for example in the device of Fig. '7, without needing a magnetic field, but by supplying the electrode P with a voltage which is negative with respect to the cathode, while the helix is supplied with a positive voltage. This system is shown in Fig. 10 which only differs from Fig. 7 by a permutation of the connections of the electrodes C, P, H. As in a tube with a positive grid, the electrons are thus reflected by the negative electrode and oscillate between same and the cathode until they are collected by the helix 1-1. This considerably heats the helix, since its elongated shape is not advantageous for carrying away heat. It is therefore preferable to use a special grid A (Fig. 11) for accelerating the movement of the electrons and to convert the helix H into a reflecting electrode by supplying it with a negative voltage. In this case, it is possible to give to the positive electrode a favorable shape for carrying the heat away. In order to make the D. C. field uniform in the axial direction between said special grid and the helix, the electrode P can be used in a similar manner to that hereinbefore described for the progressive- Wave tube with a magnetic field, but in this case said electrode would have to be supplied with. a high negative voltage. Under certain conditions, it may also be advantageous, for this tube, to place the cathode C outside and the electrode P on the axis of the system (Fig. 12), in particular in the case in which, for carrying the heat away, it is necessary to have a positive grid of the largest possible area, while the diameter of the helix must remain small.

It should also be noted that it is not essential to make the delay line of the shape of a single helix, its being possible, for example, for said line to be formed by a suitably chosen bifilary line. Similarly, it is not essential to choose a radially symmetrical shape for the electrode system.

It is possible to obtain an amplification effect with these tubes even in the case in which the period of oscillation of the electrons is substantially shorter than that of the progressive wave, which can be considered as an important advantage. Consequently, these tubes can operate with comparatively low D. C. voltages while using very high frequencies and fairly large spacings between the electrodes, since for a given spacing, the voltages are proportional to the square of the oscillation frequency of the electrons.

The electrons can only transfer their energy to the A. C. field of the progressive wave, for low A. C. voltages, in small quantities and consequently it is necessary for there to be a large number of oscillations for each electron that transfers energy. It is therefore advisable to arrange for the period of oscillation of the electrons that transfer energy to be constant, independen ly of the amplitude, and for the electrons which are not of the requisite phase (i. e. those which absorb the energy from the A. C. field) to be converted into electrons of the correct phase. These results can be obtained in the same manner as has been successfully used for stationary oscillations of similar tubes, for example by deforming the reflecting field by means of an auxiliary grid of adjustable voltage.

The fact that the tubes are symmetrical is important for the practical Working. As a result thereof an amplification can take place both for a progressive wave which is propagated from E towards S and for a wave which is propagated in the opposite direction, from S towards E. The input and the output can therefore be permuted. This may, if necessary, facilitate self-oscillation if, owing to the insuificient matching of the output apparatus at the output of the delay line, part of the energy is reflected and returns from S to E.

To decrease the danger of self-oscillation, the tubes may be rendered asymmetrical by substantially reducing the undesirable amplification in the direction S-E with respect to the desired amplification in the direction E-S, as a result of which the input and the output can no longer be permuted. A certain degree of asymmetry in the case of the tube with a magnetic field hereinbefore described is moreover already produced by the fact (not hereinbefore mentioned) that the electrons efiect a movement about the axis of the system of the tube, parallel to a convolution of the helix (see Figs. 3 and 9). This movement therefore takes place in the same direction or in the opposite direction to the direction of propagation of the progressive wave along the wire of the helix. If, for example, the said movement takes place in the same direction as the propagation of said wave from E to S, it is in the opposite direction to the direction of propagation from E to S of the reflected wave. Consequently, diiferent conditions are produced for the oscillation frequency of the electrons for the incident Wave and for the reflected wave. If this frequency is adjusted in the most advantageous manner for the progressive wave travelling from E to S, the conditions for the wave travelling in the opposite direction are less favorable. The tube is thus asymmetrical.

If this asymmetry is not sufficient, or if it does not exist from the outset, as is the case for tubes without a magnetic field, the tubes have to be rendered asymmetrical by artificial means. In principle, this result can be obtained by continuously or gradually varying the operating conditions along the tube, from E to S, in such a manner as to make them very advantageous near E for low A. C. voltages and for high voltages near S. This can be effected in practice by using various methods. For example, the distance between the convolutions may be made comparatively great near E and be successively decreased towards S. In this case, an active exchange of energy with the electrons is only obtained in the presence of fairly high A. C. voltages, for a small spacing between the convolutions. It is also possible to use a composite refleeting electrode comprising a plurality of electrodes which are aligned between E and S and operate at difierent reflecting voltages. In this case, the desired asymmetries can be obtained by gradually or continuously varying the distance between or the dimensions of the electrodes.

What I claim is:

An ultra-short wave amplifying tube, compris ing a delay line having an input end and an output end, means coupled with said input end for feeding an electromagnetic wave into the delay line, means coupled with the output end for 8 removing amplified energy from the delay line, an emissive cathode extending coaxially with respect to the delay line, a direct potential source connected to the extremities of said cathode for passing a heating current therethrough, said cathode being traversed only by said heating current, means comprising a source of voltage connected between said line and said cathode for producing a time-constant electric field therebetween, and a magnet positioned with respect to said tube so as to produce a uniform time-constant magnetic field traversing the said tube in the direction of the common axis of the cathode and the delay line, whereby an oscillatory movement between said cathode and line is imparted to electrons emitted by said cathode.

ALFRED LERBS.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 1,877,872 Hollmann Sept. 20, 1932 2,122,538 Potter July 5, 1938 2,241,976 Blewett et a1. May 13, 1941 2,300,052 Lindenblad Oct. 27, 1942 2,402,184 Samuel June 18, 1946 2,439,401 Smith Apr. 3, 1948 2,511,407 Kleen et al. June 13, 1950 2,516,944 Barnett Aug. 1, 1950 2,541,843 Tiley Feb. 13, 1951

Images