NEW INSTRUMENTS AND METHODS OF MEASUREMENT

Oscillograph Tubes for Recording Fast-Running Processes

V. I. Voznesenskii, N. V. Korotkikh,
A. V. Chernetskii, A. S. Koporskii

INTRODUCTION

One of the most important tasks of modern science is the investigation of processes that take place over extremely short intervals of time—billionths of a second and smaller fractions of a second. The study of such phenomena by means of ordinary oscillographs encounters a number of fundamental difficulties.

The point is that, when the repetition period of processes or their duration is reduced, one has to deal with entirely new phenomena that considerably complicate their indication. These include, first of all: a decrease in sensitivity and distortion of the image due to the transit-time effect in oscillograph tubes; distortion of the signal because of impedance mismatch at the points where the signal passes into the deflecting system and because of parasitic couplings between deflecting systems; a decrease in image brightness when the recording speed is increased; insufficient resolving power of ordinary tubes; insufficient bandwidth and stability of the amplifiers of ordinary oscillographs, etc. These and other limitations make it essentially impossible to use ordinary types of oscillographs at frequencies above 5–10 Mc/s and for observing pulses of duration less than a few fractions of a microsecond.

In recent years, both in our country and abroad, new types of oscillographs have been developed that have made it possible to extend considerably the range of measurements toward shorter-duration processes. This has been achieved not only, and not even so much, by improving old methods, but mainly through the development of fundamentally new methods of oscillography*). Significant changes have also been introduced into amplifier circuits, sweep generators, and various auxiliary devices of oscillographs.

In this review, covering the last decade, the principal methods of oscillography of fast-running processes are considered, as well as certain issues characteristic of high-speed cathode-ray tubes, such as obtaining a thin electron beam, post-acceleration, etc.

*) Since the method of deflecting the electron beam largely determines the high-frequency measurement limit of oscillographs and imposes certain requirements on the entire tube design and on the circuits of the main units of oscillographs, by the method of oscillography here is meant mainly the method of deflection.

§ 1. METHODS OF HIGH-SPEED OSCILLOGRAPHY

Deflecting systems. Limitations of ordinary deflecting systems at UHF

Before proceeding to consider new methods for observing rapidly varying processes, it is useful to dwell in somewhat greater detail on the limitations that ordinary deflecting systems have at UHF1.

The deflection system of ordinary tubes consists of two pairs of parallel plates arranged in mutually perpendicular planes2. Such a system is shown in Fig. 1. The signal under investigation is applied to one pair of plates; the other pair of plates is used to sweep this signal in time.

Fig. 1. Ordinary electrostatic deflecting system.

At low frequencies of the voltage under investigation (or for long pulse durations), when the transit time of the electrons between the deflecting plates is much smaller than the period of variation of the voltage on the plates, the deflection of the beam is strictly proportional to the potential difference applied to them. The magnitude of the deflection in this case can be determined from the expression

\[ A_0=\frac{lL}{2dV}\,U_0\cos \omega t, \]

where \(U_0\cos \omega t\) is the voltage under study (V), \(l\) is the length of the plates (cm), \(L\) is the distance between the plates (cm), \(V\) is the beam potential (V) (accelerating voltage), and \(d\) is the distance between the deflecting plates.

It follows from this expression that the magnitude of the deflection varies in time according to the same law as the field, without phase shift. The deflection sensitivity, or the magnitude of deflection per unit deflecting voltage,

\[ a_0=\frac{lL}{2dV}, \]

does not depend on frequency.

At high frequencies, when the electron transit time is comparable with the period of variation of the deflecting field, the magnitude of the deflection varies according to a more complicated law:

\[ A(\omega)=\frac{lL}{2dV}\,U_0\cos\left(\omega t_0+\frac{\omega\Delta t}{2}\right)\cdot \frac{\sin\left(\frac{\omega\Delta t}{2}\right)}{\frac{\omega\Delta t}{2}}, \]

where \(\Delta t\) is the transit time of the electrons through the field of the deflecting plates.

In this case a phase shift is observed between the deflection and the deflecting voltage, equal to

\[ \frac{\omega\Delta t}{2}. \]

The ratio of the deflection amplitude in rapidly varying fields to the deflection amplitude in slowly varying fields is

\[ \frac{A(\omega)}{A_0}= \frac{\sin\left(\frac{\omega\Delta t}{2}\right)} {\left(\frac{\omega\Delta t}{2}\right)}. \]

The relation between the static and dynamic sensitivities has the form

\[ a(\omega)=a_0\,\frac{\sin\left(\frac{\omega \Delta t}{2}\right)}{\left(\frac{\omega \Delta t}{2}\right)}, \]

i.e., the deflection sensitivity is proportional to the ratio of the sine of half the angle of transit through the deflecting plates to half this same angle. Thus, for example, at a transit angle of \(0.794\) radian the ratio of the dynamic sensitivity to the static sensitivity is \(0.9\); at a transit angle of \(2.78\) radians this ratio is approximately \(0.7\), and at a transit angle equal to \(2\pi\) the ratio becomes zero.

In Fig. 2 the dependence of the deflection sensitivity on frequency is presented, calculated by Holman for the case \(V=1000\) V and a length of the deflecting plates \((l=2\ \mathrm{cm})^8\). It is evident from the figure that, for the given dimensions of the deflecting plates, the deflection sensitivity drops sharply beginning with frequencies of the order of \(10^8\ \mathrm{Hz}\). At a frequency of \(6\cdot 10^8\ \mathrm{Hz}\) the sensitivity falls to \(50\%\) of its initial value, and at a frequency of \(1\cdot 10^9\ \mathrm{Hz}\) it falls to zero. It must be borne in mind that the calculation was made without taking account of the fringe field. When this factor is taken into account, the frequency characteristic of the system deteriorates.

Fig. 2. Dependence of the dynamic sensitivity of an ordinary deflecting system on frequency.

Thus, even for observing harmonic oscillations at frequencies above \(1\cdot 10^8\ \mathrm{Hz}\), serious difficulties arise.

The situation is still worse with nonharmonic and especially pulsed signals having in their spectrum components of very high frequencies. In order to obtain an undistorted image of a pulse with a rise time \(\tau_{\mathrm{f}}\), it is necessary to reproduce all harmonic components lying in the frequency band at least up to

\[ \frac{1}{\tau_{\mathrm{f}}}, \]

and in order to reproduce a signal of exponential form it is necessary that the time constant of the exponential be at least six times greater than the transit time of the electrons between the deflecting plates. If these conditions are not fulfilled, considerable distortions of the image arise because of the different deflection sensitivity for different harmonics.

Another limitation that must be encountered when ordinary oscillographs are used for observing rapidly occurring processes is the distortion arising from the considerable capacitances between the deflecting plates and from parasitic couplings between the two pairs of plates. Ordinary oscillographic tubes have rather large values of capacitance between the deflecting plates (of the order of \(5\)–\(10\ \mathrm{pF}\)). With such capacitance values, at high frequencies the charging time begins to have an effect; it may be approximately estimated with the aid of the “charging constant,” equal to the product of the capacitance of the plates and the internal resistance of the signal source*).

\[ \text{*) The full charging time of a capacitor is } \sim 5RC. \]

The charging time for ordinary tubes is comparable with the duration of the pulses, which leads to lengthening of their fronts.

A decrease in the capacitive resistance of the deflecting plates with increasing frequency leads to the voltage across them during the charging time becoming less than that at the source. This, in turn, leads to a reduction in sensitivity at high frequencies for harmonic signals and to distortion of the image of signals containing higher harmonics.

As a result of parasitic couplings between two pairs of deflecting plates and their leads, modulation is observed of the high-frequency voltage applied to one pair of plates by another high-frequency voltage applied to the other pair of plates. In this case, the velocity of motion of the beam in the horizontal direction (the time axis), during the action on the vertical plates of the leading edge of the signal being studied, may decrease, while during the action of the trailing edge, on the contrary, it increases.

Fig. 3. Distortions of the image of a pulse due to parasitic couplings between the deflecting plates.

The typical picture obtained in this case on the screen is shown in Fig. 3.

Serious difficulties also arise because of impedance mismatching at the points where the signal passes to the deflecting plates. The reflected waves that arise in this case are superimposed on the signal being studied, thereby causing distortion of its image. Matching the deflecting system with the generator is also important for obtaining signals stable in frequency and power.

Questions such as obtaining a thin electron beam with a high current density and efficient phosphors are also of very great importance for high-speed oscillography.

From what has been said it is clear that the high-frequency limit of cathode-ray tubes will be determined by how completely all these problems are solved.

To reduce the transit time, one can reduce the length of the deflecting plates and increase the velocity of the beam electrons; however, this leads to a decrease in sensitivity. Therefore, when designing high-speed tubes one has to make a certain compromise between the need to have a small transit time and the need to have sufficient deflection sensitivity.

The use of high accelerating voltages is also dictated by the need to have a high recording speed. Thus, for example, with a fivefold increase in the accelerating voltage, an increase in the maximum recording speed by a factor of 15–20 is achieved. In most modern high-speed oscillographic tubes, in order to increase the energy of the electrons, so-called post-acceleration is used, i.e., additional acceleration of the beam electrons after they have passed through the deflecting system. With the aid of post-acceleration in the tube it is possible to combine such contradictory conditions as high electron energy and, consequently, high recording speed, with considerable deflection sensitivity. Reduction of capacitance and parasitic couplings between the plates and the leads is achieved by reducing the length of the plates, shortening the leads and sealing them not through a common base but directly into the bulb of the tube, by thorough shielding of one pair of plates from the other, etc. These condi-

improvements make it possible to considerably expand the frequency range of measurement of modern cathode-ray tubes.

Let us consider several models of high-speed tubes with a deflecting system in the form of plates. Typical tubes of this kind are the GEC tubes of types 908BCC and 1608BCCA, described in the literature9, 10. Photographs of these tubes are shown in Fig. 4. The first tube operates at an accelerating voltage of 10 kV and has no post-acceleration. The maximum writing speed achieved in this tube is about \(3 \cdot 10^{9}\ \mathrm{cm/sec}\). The spot diameter on the screen is \(0.2 \div 0.25\ \mathrm{mm}\). The deflection sensitivity is \(4 \div 4.5 \cdot 10^{-2}\ \mathrm{mm/V}\). Such a low sensitivity makes it impossible to use this tube with signals smaller than \(100\ \mathrm{V}\). The sweep voltage of the tube is \(2.5\ \mathrm{kV}\). According to calculations, such a

Fig. 4. a) GEC cathode-ray tube type 908BCC. b) GEC cathode-ray tube type 1608BCCA with post-acceleration.

tube can operate satisfactorily up to frequencies of the order of \(200 \div 300\ \mathrm{MHz}\).

The second tube is designed with post-acceleration. The electrons of the beam are first accelerated by a potential of \(3\ \mathrm{kV}\), and then, after passing through the deflection system, are additionally accelerated by means of three post-acceleration rings to a potential of \(7\ \mathrm{kV}\). The sensitivity of the tube is \(1.5 \cdot 10^{-1}\ \mathrm{mm/V}\). This makes it possible to observe signals with a voltage amplitude of \(40\ \mathrm{V}\). The sweep voltage of the tube is \(700\ \mathrm{V}\). The spot diameter on the screen is about \(0.5\ \mathrm{mm}\). However, this tube has a lower writing speed \((1 \cdot 10^{9}\ \mathrm{cm/sec})\) and a lower high-frequency limit \((100—150\ \mathrm{MHz})\).

The examples given show that, by improving the design and operating conditions of oscilloscope tubes with a deflecting system in the form of plates, it is possible to achieve a considerable expansion of the measurement range for fast processes. However, with

In evaluating the data presented, it should be remembered that they were obtained without taking into account capacitance and other factors that greatly reduce the effectiveness of their application at high frequencies. The low sensitivity of these tubes imposes very stringent requirements on the amplifiers and sweep generators: a large gain factor, high output voltages, and sufficient bandwidth are required. In addition, in these designs the question of matching the signal sources with the deflecting system remains unresolved. Taking all this into account, it may be said that the high-frequency limit for tubes with a deflecting system in the form of parallel plates is frequencies of the order of 600–800 MHz.

Let us now turn to consideration of cathode-ray tubes that were developed in recent years specifically for oscillography of rapidly occurring processes. Undoubtedly, all the difficulties and limitations noted in this paragraph were taken into account in the creation of these tubes and, to one degree or another, were overcome, which made it possible to considerably widen the range of transmitted frequencies, increase the sensitivity of the tubes, etc.

§ 2. CATHODE-RAY TUBES WITH A DEFLECTING SYSTEM IN THE FORM OF A TWO-WIRE LINE

The effort to reduce to a minimum the transit time of electrons through the deflecting system led to the idea of using a two-wire line for deflecting the electron beam (Fig. 5). In addition to reducing the transit time, this achieves a considerable reduction in capacitance and an improvement in matching of the deflecting system with the signal and sweep source, since such a system represents, for a high-frequency wave, a distributed impedance in the form of a two-wire transmission line. This makes it possible to connect one end of the deflecting system to the test apparatus and to load the other end with the characteristic resistance. In addition, such a system can easily be tuned to resonance, which makes it possible to obtain considerable deflecting voltages. Naturally, the deflection sensitivity of such a system is low.

Fig. 5. Diagram of a two-wire deflecting system.

Its estimate may be made by the formula¹:

\[ \theta = \frac{U\pi}{2V\,\operatorname{Arch}\left(\frac{D}{d}\right)} \simeq \frac{U\pi}{2V\ln\left(\frac{2D}{d}\right)}, \]

where \(U\) is the deflecting voltage \((\mathrm{V})\), \(V\) is the velocity of the electrons \((\mathrm{V})\), \(D\) is the distance between the centers of the lines, and \(r\) is the radius of the wire lines.

One of the tubes with a deflecting system in the form of a two-wire line¹² operated at an accelerating voltage of 25 kV. In this case, in the central part of the screen (approximately 4 cm in diameter), a maximum recording speed of \(4 \cdot 10^{10}\) cm/sec was achieved with a spot size on the screen of 0.2 mm. The deflection sensitivity for direct voltage was only \(7 \cdot 10^{-3}\) mm/V. At a frequency of 10,000 MHz the sensitivity fell to 30% of its initial value. In this tube an aluminized screen with an extremely effective phosphor was used. Of interest is the application of a tube with a deflecting system in the form of two pairs of two-wire lines as an ultrafast beam analyzer¹³, with the aid of which studies were made of the velocity and density-

modulation of the electron beam, modulated at a frequency of 3000 MHz, at the output of a klystron or a traveling-wave tube. The deflecting system was tuned to resonance at the frequency that modulates the beam. Since electrons having different velocities are deflected differently, a complete picture of the distribution of velocities and electron density in the beam is obtained on the screen. A circular sweep was used in the work. A resolution of 30 V was achieved, and the authors point out the possibility of reducing this value.

Oscillographs with a two-wire deflecting system are an attempt to solve the problem of observing rapidly occurring processes. However, because of their low sensitivity they have not come into wide use. The field of their application is naturally limited to those rapidly occurring processes whose voltages are on the order of several kilovolts.

§ 3. ELECTRON-BEAM TUBES FOR THE INVESTIGATION OF HIGH-POWER RAPIDLY OCCURRING PHENOMENA

Oscillographing very-high-frequency processes of high power has a number of characteristic features that impose very stringent requirements on the design of tubes intended for these purposes. Such tubes must, first, be fast and sufficiently wide-band, and, second, must allow operation with high-power signals. It is therefore natural that the study of high-power very-high-frequency processes with ordinary fast tubes encounters serious difficulties. Thus, for example, the use of special voltage dividers, necessary in this case, and of other auxiliary circuits is inevitably associated with noticeable distortion of the signal under investigation. Moreover, these distortions are especially large when oscillographing short-duration transient processes.

Recently descriptions have appeared of several types of electron-beam tubes intended specifically for observing short-duration high-power transient processes.

One such tube^14 had a deflecting system consisting of a cylindrical electrode 5–6 mm in diameter, placed in a similarly shaped screen with a slot, and located opposite the slot of another electrode. The screen and the electrode located opposite the slot of the screen were grounded. Deflection of the electron beam was produced by the field between the inner cylindrical electrode and the grounded electrode. The deflecting system was designed for signals with an amplitude of about 100 kV. The tube operated at an accelerating voltage of 18 kV. Focusing of the electron beam was carried out by means of a magnetic lens. Unfortunately, the authors of the tube do not indicate the dimensions of the deflecting system, the recording speed, or the diameter of the spot on the screen of this tube.

Often tubes of this type^15 are constructed with a double deflecting system: one of them has low sensitivity, the other normal, thereby making it possible to observe both high- and low-voltage signals.

An interesting method of oscillographing short-duration high-power transient processes was developed by A. M. Chernushenko.^16 According to this method, deflection of the electron beam is produced by means of a cavity resonator used as the deflecting system. The resonator is formed by two cylindrical cavities connected to each other by a slot. The walls of this slot form the lumped capacitance of the resonator and are used as deflecting plates.

The cylindrical cavities forming the resonator inductance are connected with the microwave source under study and with the load. Energy is supplied to the deflecting system from a coaxial line, whose outer conductor has a capacitive coupling with the body of the resonator, and whose inner conductor has an inductive coupling with the loop.

A tube with such a deflecting system was designed for investigating ultrahigh-frequency oscillations of pulsed magnetrons in the decimeter range. In the range 8–12 cm the deflecting system of this tube could transmit not less than 90% of the applied high-frequency energy. The deflection sensitivity was 2 V per line width. Figure 6 shows a general view of the SVCh-1 tube, designed by A. M. Chernushenko.

Fig. 6. General view of the SVCh-1 oscillographic tube.

The described method of oscillographing ultrafast processes has a number of advantages. These include the possibility of concentrating the main part of the energy in the region of deflection of the electron beam, large amplification of the signals being investigated, the relative simplicity of the tube design, the absence of complicated vacuum seals between glass and metal, etc. A substantial drawback is the comparatively small bandwidth of the deflecting device. This limits the application of such tubes to processes with a small spectral width.

§ 4. MICRO-OSCILLOGRAPHIC TUBES

The distinguishing features of micro-oscillographic tubes are: small dimensions of the deflecting plates (~mm), high accelerating potentials (~10 kV), an extremely thin electron beam (~10^-2 mm), and small image dimensions requiring additional magnification. The image is often recorded on photographic plates; in this case the tube is made demountable, which necessitates the use of pumping.

Reducing the length of the deflecting plates at high electron velocities in the beam makes it possible to reduce considerably the time of flight of the electrons through the deflecting system and to reduce the capacitance between the plates. However, this leads to a loss in sensitivity. The image—

... the image on the screen becomes small, and additional magnification is required in order to observe it. In addition, to increase the resolving power of the instrument it is necessary to have a thin electron beam and a fine-grained screen.

The use of a sensitive photographic plate as the indicator makes it possible to obtain a sharp image even at very high recording speeds \((10^9\ \text{cm/sec})\). Replacement of the plates and restoration of the working pressure \((10^{-4}—10^{-5}\ \text{mm Hg})\) usually takes no more than 10–15 minutes. About 100 oscillograms can be recorded on a single photographic plate. In most cases the electron beam is focused with the aid of carefully adjusted magnetic lenses.

Fig. 7. Diagram of a microoscillographic tube.

Figure 7 shows a typical diagram of a microoscillographic tube. Such a tube consists of an electron source (a three-electrode system is shown, consisting of needle cathode 1, control electrode 2, and anode 3), a magnetic reducing lens 4, the main magnetic lens 5, a vertical deflection system 6, a horizontal deflection system 7, and screen 8. For adjustment and tuning of the instrument there is usually a removable luminescent screen, the image on which can be observed through a special “window” in the tube. Diaphragms are usually placed between the poles of the magnetic lenses.

Because it is possible to enlarge the image obtained on the photographic plate, many types of microoscillographs make it possible to work with signals having amplitudes on the order of several volts.

Thus, in the first microoscillographic tube, described by Ardenne in 1939,^17 by means of careful adjustment of the magnetic focusing system a spot approximately \(10^{-3}\ \text{mm}\) in diameter (!) was obtained. The dimensions of the deflecting plates were reduced by a factor of 10. As a result, an oscillogram 100 times smaller than usual was obtained.

In later designs Ardenne used a combined focusing system consisting of electric and magnetic lenses.^18

Another design of a microoscillograph had deflecting plates measuring \(5 \times 4\ \text{mm}\), with a spacing between them of \(3.6\ \text{mm}\). Each pair of plates was carefully shielded, which made it possible to reduce parasitic coupling to a minimum. To improve focusing of the electron beam, a special focusing lens was used, placed between the deflection system and the screen. The sensitivity of the deflection system of this tube was \(1 \cdot 10^{-3}\ \text{mm/V}\). The spot size on the screen was \(10^{-2}\ \text{mm}\). With an accelerating voltage of \(50\ \text{kV}\), the maximum recording speed was \(2 \cdot 10^9\ \text{cm/sec}\). Such a tube, according to calculation, could be used up to frequencies on the order of \(10\,000\ \text{MHz}\). However, oscillograms of single oscillations obtained by the author of this design at a frequency of \(10\,000\ \text{MHz}\) proved unsatisfactory. Apparently, at such high frequencies such phenomena as local—

... resonances, radiation fields, and the superposition of focusing and deflecting fields.

In another micro-oscillograph, described by Fert et al.^21, magnetic focusing of the beam was also used. This tube had an accelerating voltage of 50 to 60 kV. A maximum recording speed of \(6 \cdot 10^9\) cm/sec was achieved.

The tube described by Pierce^22 had a deflecting system in which the plates formed lumped capacitances of a low-frequency filter. The plates were connected by inductances. The system thus obtained had a low-frequency impedance of 75 ohms. This system was designed for a limiting frequency of the order of 1300 MHz; however, owing to the variation of the characteristic impedance with frequency and insufficiently accurate selection of the components, this limit was reduced to a frequency of 500 MHz. The sensitivity of the tube was \(7.7 \cdot 10^{-1}\) mm/V; the spot diameter was \(2 \cdot 10^{-2}\) mm. The low accelerating voltage (2 kV) did not permit this micro-oscillograph to be used for recording nonstationary processes.

The examples given show that micro-oscillography makes it possible to extend the frequency range of measurement up to several thousand MHz. In the micro-oscillograph it has been possible, to a considerable extent, to smooth out the contradiction between the need to reduce the length of the deflecting plates and to increase the velocity of the electrons in the beam in order to reduce the transit time, and the need to have sufficient deflection sensitivity. Photographing the image on a sensitive film with subsequent enlargement by 50–100 times makes it possible to work with small images. The image obtained is sufficiently sharp owing to the very fine focusing of the electron beam and to the fine-grained screen. This makes it possible to measure very weak signals, starting from fractions of a volt. The usual magnitude of the required signal does not exceed 10 V. In addition, it should be taken into account that recording speeds of more than \(10^9\) cm/sec can be achieved in micro-oscillographs. All this makes these instruments indispensable for the study of many rapidly varying processes, including single pulses.

The disadvantages of micro-oscillographs include the complexity of their design and manufacturing technology, their bulkiness, and the interruption of operation when films are changed.

§ 5. TUBES WITH A TRAVELING-WAVE DEFLECTING SYSTEM

Increases in sensitivity with large bandwidth have been achieved in tubes with a traveling-wave type deflecting system. Figure 8 shows a diagram of such a deflecting system. The electron beam propagates along the axis \(AB\) and is deflected by the electric field of two curved parallel conductors, along which an electromagnetic wave propagates in the direction of motion of the electrons. The phase velocity of the electromagnetic wave depends only on the geometry of the system and can be made equal to the velocity of motion of the electrons. In this case each electron passing through the deflecting device is subjected to a deflecting voltage constant in magnitude and direction. Thus, it becomes possible to achieve a significant increase in the sensitivity of the deflecting system through multiple interaction of the beam electrons with the electric field of the process under investigation. The transit time of such a system (and, consequently, the relative magnitude of signal distortions due to the transit-time factor) is determined by the transit time of the electrons through one loop and can be made sufficiently small.

In addition, in tubes with a traveling-wave deflection system it is possible to achieve good matching with the signal source. These merits of tubes with a traveling-wave deflection system make them very promising for the study of rapidly varying processes.

In recent years reports have appeared on several designs of such tubes. One of the first traveling-wave tubes[^23] had a deflection system in the form of two parallel wire loops (Fig. 8). The tube design was in the nature of an experimental development. According to the authors’ calculation, the tube could operate without any appreciable reduction in sensitivity up to 30,000 MHz. However, the calculation did not take into account radiation, reflection in the glass, finite-line-length effects, the electron transit time through one loop, etc. The tube had an accelerating potential of 2.5 kV and a sensitivity of 0.1 mm/V. With its aid, observations were made of Lissajous figures and frequency analysis of continuous waves by Holmen’s method of inversion spectrograms.

Fig. 8. Diagram of the deflection system of a traveling-wave tube.

A tube of a later design[^24] was adapted for mass production. It was intended for recording aperiodic processes of millimicrosecond duration. The deflection system consisted of a strip semicircular spiral with a flat side, placed in a screen of similar shape. The electron beam passed between the flat parts of the screen and the spiral. Owing to the fact that the spiral along its entire length was at a constant distance from the screen and was mounted from strip of constant width, the deflection system acted as a line with a definite inductance and capacitance. In this tube the deflection system was designed for a wave impedance of 50 ohms (spiral strip width 1 cm; distance between spiral and screen 0.2 cm). With an accelerating voltage of 10 kV the transit time was \(0.17 \cdot 10^{-9}\) sec. The deflection sensitivity was \(2.5 \cdot 10^{-1}\) mm/V. A five-percent decrease in sensitivity was observed at a frequency of 1000 MHz. With the aid of five post-acceleration rings the beam potential was raised to 35 kV, which made it possible to obtain a maximum recording speed of \(1.5 \cdot 10^9\) cm/sec. The spot diameter on the screen was 0.15 mm. The tube used a screen with a very efficient phosphor, coated with a thin mirror layer of aluminum.

In the Soviet Union, a tube of similar design was first made in 1953 by S. G. Konstantinov during his diploma design work. In work[^25] a calculation is given of the sensitivity of a traveling-wave deflection system, and the frequency characteristics of the tube are presented up to frequencies of the order of 3000 MHz.

In tubes designed by A. M. Chernushenko and P. A. Tarasov[^26], two varieties of traveling-wave deflection systems were used, with spiral and flat loops. The deflection system was calculated for a characteristic impedance of 100 ohms. The accelerating voltage in these tubes was 10 kV. With the aid of four post-acceleration rings the beam potential was raised to 35 kV. In this case the maximum recording speed was \(2 \cdot 10^{10}\) cm/sec. It should be noted that in these tubes a metallized screen with a very efficient ...

luminescent screen, which, all other conditions being equal, makes it possible to obtain a spot brightness four times greater than with the luminescent screens used previously.

The deflection sensitivity is about \(0.5\ \text{mm/V}\). The full deflection along the phenomenon axis is \(2\ \text{cm}\), and along the time axis \(10\ \text{cm}\). For these tubes a projector was specially developed, in which the cathode lens creates a parallel beam of electrons. Electrostatic focusing was used in the tubes. The diameter of the spot on the screen is about \(0.2\ \text{mm}\). According to the designers, the tubes can be used effectively up to frequencies of the order of \(10\,000\ \text{Mc/s}\).

An original method for oscillographing rapidly occurring processes was developed at the Molotov Moscow Power Engineering Institute[^27]. Deflection of the beam in the oscillographic tube is carried out by means of a two-conductor (two-start) helical line whose axis coincides with the axis of the tube (Fig. 9).

Fig. 9. Diagram of a self-sweeping deflecting system.

This helical deflecting system is fed from a two-conductor, preferably symmetrical, feeder. The other end of the helix may be open or loaded with a matched resistance.

When an electromagnetic wave passes through such a helical deflecting system, the electric field rotates. As a result, the force acting on the electrons in each subsequent section of the deflecting system is directed at a certain angle to the direction of the transverse velocity of the electrons acquired by them in the preceding section. The speed of rotation of the field relative to the electrons of the beam is inversely proportional to the pitch of the helix and depends on the beam velocity, while the total angle of rotation of the field is determined by the length and pitch of the helix.

Such a deflecting system can be placed in a longitudinal magnetic field. In this case it is also necessary to take into account the rotation of the transverse-velocity vector of the electrons through an angle proportional to the cyclotron frequency and to the time of travel of the electrons from one section of the system to another.

An interesting variant of the deflecting system is a two-conductor straight transmission line, parallel to the electron beam, placed in a longitudinal magnetic field. In this case the deflecting system has greater bandwidth and less dispersion. The speed of rotation of the electric field relative to the electrons of the beam is determined by the cyclotron frequency

\[ \omega_0=\frac{eB}{m}, \]

and the total angle of rotation is a function of the magnitude of the magnetic field and the transit time.

It should be noted, however, that the use of magnetic coils considerably complicates the design of the oscillograph. Serious difficulties also arise with focusing the electron beam and adjusting the magnetic field.

Tubes in which deflection of the electron beam is carried out by means of a two-conductor helical line without the use of magnetic coils have proved more promising.

A characteristic feature of a tube with a deflecting system of this type is that, unlike other deflection systems, it can operate only when the period of variation of the process under investigation is less than or equal to the transit time, i.e. such tubes can in fact operate only as self-sweeping tubes.

be used only for indicating oscillations with a frequency greater than 100 Mc/s and pulses shorter than 10 mµs. However, when small accelerating voltages are used, it is possible to observe longer processes. In this case a sufficient recording speed can be ensured by means of post-acceleration. A great advantage of this tube is that its deflection system is self-sweeping. Thus the need for a sweep generator is eliminated. In addition, the use of tubes with a central collector that collects the electrons of the undisturbed beam makes it possible to eliminate all auxiliary circuits from the oscilloscope. In this case the oscilloscope consists only of the power-supply rectifiers and the tube.

An important feature of this method is the possibility of bringing the deflection sensitivity up to the sensitivity of the lowest-voltage cathode-ray tubes with an ordinary deflecting system.

A disadvantage of the method is the complicated form of the oscillograms. If in ordinary tubes the signal is recorded in rectangular coordinates, then in the present case the relation between the quantity characterizing the phenomenon and time proves to be considerably more complex. For example, when oscillographing UHF oscillations the trace of the beam is an ellipse, while modulated oscillations are obtained in the form of a ring whose thickness depends on the modulation depth. Thus, decoding of the oscillograms is necessary.

Holmen[^29] proposed a deflecting system consisting of three pairs of plates. The electron transit time through one pair of plates is chosen so that it is equal to one half-period of the sinusoidal wave. Thus the total deflection of the electrons is equal to the sum of the deflections acquired by them in passing through each pair of plates. It should be noted, however, that the frequency region in which such a deflecting device can be used is rather small. Therefore the use of such a system is limited.

As the examples given show, cathode-ray tubes with a traveling-wave deflecting system have made it possible to extend the frequency range of oscillography up to frequencies of the order of 5–10 thousand Mc/s with satisfactory sensitivity (of the order of 0.2–0.5 mm/V) in good matching.

At present, traveling-wave tubes are becoming a powerful tool for the investigation of rapidly occurring processes. The disadvantages of such tubes include the complexity of the design both of the deflecting system and of the tube as a whole, and signal distortions due to the finite dimensions of the deflecting system and the transit-time factor.

§ 6. INVESTIGATION OF ULTRASHORT ELECTRON BUNCHES

Somewhat apart from cathode-ray tubes intended for indicating electromagnetic processes are tubes intended for the direct observation of very short electron bunches. Their appearance is connected with the development of millimeter-wave technology, nuclear physics, and so forth. The point is that the task of obtaining millimeter-wave, millimicrosecond, and shorter pulses consists mainly in developing methods for obtaining sufficiently short electron bunches of high density. The theoretical analysis of many questions of this problem is extremely difficult. The experimental study of electron bunches, including such questions as the measurement of the duration, observation of the shape and density of bunches, and the study of phenomena-

…associated with space charge, etc., requires the creation of a very delicate measurement technique.

To measure the duration of electron bunches it is convenient to use the circular-sweep method, which makes it possible to increase the time scale considerably. This type of sweep is obtained by applying to the deflecting system two sinusoidal voltages equal in frequency and amplitude, but shifted in phase by \(90^\circ\). When a continuous beam is swept, a circle is obtained on the screen; when a beam cut into bunches is swept, an arc of a circle is obtained, the length of which is proportional to the duration of the bunch.

\[ t=\frac{T}{2\pi R}\,l=\frac{T}{2\pi}\,\alpha, \]

where \(T\) is the period of the sweeping voltage, \(l\) is the length of the arc, \(\alpha\) is the length of the arc in radians, and \(2\pi R\) is the circumference of the sweep.

For measuring bunches with durations of \(10^{-9}\) sec and less, tubes with a deflecting system in the form of two-wire lines are usually used \(^{30}\). With such tubes, by means of a circular sweep it is possible to measure electron bunches with durations down to \(10^{-11}\)—\(10^{-12}\) sec.

The drawback of this method is the impossibility of observing the shape of the bunches.

A very attractive method is the so-called “crossed-beams method” \(^{31}\): the study of electron bunches of high current density is carried out by sweeping a transverse probing beam and observing the Coulomb spreading on the screen of the tube. In Fig. 10 a diagram is presented of the tube with which this idea was tested.

Fig. 10. Diagram of a tube with crossed beams.

Under the experimental conditions the pulsed beam had a bunch duration of about \(2\) μsec at an accelerating voltage of \(8\) kV and a current of \(0.115\) mA. The probing beam had a potential of \(1.45\) kV.

Comparison of the images obtained by means of the crossed-beams method and by means of a synchroscope gives good agreement.

The vertical deflection was about \(5\) mm. The method \(^{32}\) makes it possible to determine the angle of spreading.

Thus, the crossed-beams method makes it possible to obtain a considerably more complete picture of the electron bunches under study than do other methods. However, the assertion of the authors of this method that the crossed-beams method makes it possible to study bunches with durations less than \(10^{-12}\) sec is doubtful. With such short bunches their charge will be too small to obtain sufficient deflection of the probing beam. In addition, one must not forget the distortions associated with the time of flight of the electrons of the probing beam through the investigated

Table 1

Table II

Parameters of Several Electron-Beam Tubes

clump. Thus, for a final evaluation of the method of intersecting beams, additional experimental work is necessary.

Concluding the consideration of methods of ultrahigh-speed oscillography, it should be noted that in this field of technology considerable progress has recently been achieved. With the aid of new methods it has proved possible to extend the frequency range of oscillography to frequencies on the order of 3–5 thousand megacycles and even higher. Table II gives the principal characteristics of several high-speed cathode-ray tubes.

OTHER ELEMENTS AND CHARACTERISTICS OF HIGH-SPEED TUBES

As has already been said, the development of oscilloscope tubes for recording fast processes proceeded mainly along the line of improving the signal deflection system. In this connection, primary attention was given to weakening as much as possible the effect of the electron transit time in the deflection system and, thereby, to raising the upper limit of its frequency characteristic. In tubes with ordinary deflection systems in the form of plates, it proved possible to reduce the electron transit time by greatly increasing the accelerating voltage. However, this led to a sharp decrease in the sensitivity of the tubes.

The idea underlying traveling-wave deflection systems was that of prolonged interaction between the electron beam and the deflecting field. Therefore, in such tubes, along with an improvement of the frequency characteristic, sufficient deflection sensitivity is also retained.

It should be noted that, for high-speed oscillographic tubes, deflection sensitivity expressed in mm/V is insufficient for determining the quality of the tube. Indeed, for the same deflection sensitivity, a tube having a smaller spot diameter on the screen will make it possible to observe signals of smaller amplitude. Therefore the concept of relative sensitivity is introduced; it indicates the magnitude of the deflecting voltage in volts required to move the spot on the screen by a distance equal to the spot diameter. The higher the relative deflection sensitivity, the better the resolving power of the tube and the accuracy of measurement of the image on the tube screen. Thus, the resolving power of the tube can be increased by reducing the spot diameter. In modern high-speed oscillographic tubes, an increase in resolving power is achieved both by increasing the deflection sensitivity and by reducing the spot size. A number of works33, 34, 36 have been devoted to the latter problem, in which the influence of various factors on the spot diameter is investigated. Among the most significant of these may be counted the electron-optical properties of the gun, the mutual repulsion of electrons in the beam, and the natural spread of the initial velocities of the electrons upon emission from the cathode. Let us now consider the influence of the listed factors on the spot size.

§ 7. SPOT DIAMETER

Electron gun. The gun must ensure the production, in the plane of the screen, of a small transverse beam-section diameter at the greatest possible beam current. Typical designs of guns with electrostatic focusing are shown schematically in Fig. 11. Fig. 12 presents a gun with magnetic focusing. From the figures given it is evident that the gun is a two-lens sy—

Fig. 11. Designs of projectors with electrostatic focusing.

Labels in the figure:

• a)
• Modulator
• First anode
• Second anode
• Cathode
• Beam crossover
• Diaphragm for trapping secondary elements

• Limiting diaphragm

• b)

• Accelerating electrode
• Modulator
• Limiting diaphragm
• First anode
• Second anode
• Cathode
• Beam crossover

• Diaphragms for trapping secondary elements

• c)

• Accelerating electrode
• Modulator
• First anode
• Second anode
• Cathode
• Beam crossover
• Limiting diaphragm

Fig. 12. The simplest projector with magnetic focusing.

Labels in the figure:

• Modulator
• Anode
• Focusing coil
• Cathode

system. The first, cathode, lens consists of the cathode, the modulator, and the first anode or accelerating electrode. Lenses of this kind in electron optics are called immersion objectives. The electrons emitted from the cathode are collected by the cathode lens into a converging beam (see Fig. 13). The image of the cathode surface is formed in the section \(a'—b'\), and the smallest cross section of the beam is in the section \(A—B\). The second lens, which may be either magnetic or electrostatic, forms on the screen an image of the beam crossover. Consequently, the cathode lens must be designed so that it gives the minimum dimensions of the crossover.

Fig. 13. Formation of the beam in an electron projector.

The projector designs shown in Figs. 11,b and 11,c have a smaller beam-crossover diameter than the design of Fig. 11,a, owing to the presence of the accelerating electrode, whose potential is higher than the potential of the first anode.

In choosing the configuration of the lens electrodes, one should take into account the need to reduce spherical aberration. It has been shown experimentally that if the second lens is formed by two cylinders, the spherical aberration will be minimal when the diameters of the cylinders are equal \(^{36,37}\).

In addition, it has been found that the spherical aberration of such a lens becomes negligibly small if the diameters of the cylinders are approximately 8 times greater than the diameter of the limiting diaphragm. The need to limit the beam by diaphragms is caused by the desire to reduce its defocusing during deflection. However, the introduction of limiting diaphragms leads to the appearance of secondary electrons, which are then accelerated and form an unfocused flux. Reaching the screen, these electrons produce undesirable illumination of it, thereby reducing the contrast of the image. To trap the secondary electrons, diaphragms are installed, usually located in the first anode (Figs. 11,a and 11,b). In the projector shown in Fig. 11,c, there are no such diaphragms, since in this case the electron beam is limited only by electrodes having the highest potentials (the accelerating electrode and the second anode), and therefore the secondary electrons knocked out of these electrodes return back to them and cannot create scattered rays.

Magnetic focusing makes it possible to obtain a smaller spot diameter. The spherical aberration of such a lens is smaller than that of an electrostatic one, owing to the large diameter of the focusing coil, which is located outside the tube.

Mutual repulsion of electrons in the beam. Mutual repulsion of electrons in the beam limits the possibilities of increasing the current density in the beam and reducing the beam cross section. The greater the beam current density, the stronger the repulsion and the larger the diameter of the beam cross section and, consequently, the spot diameter. Therefore, the attempt to increase the brightness of the glow by increasing the beam current density leads to an increase in the spot. In each practical case a compromise solution must be found.

A theoretical analysis of this effect¹ shows that, in order to reduce the spot diameter while taking into account the mutual repulsion of electrons, it is necessary: 1) to increase the convergence angle of the beam in the focus of the projector (angle \(\Theta\) in Fig. 13); 2) to increase the voltage of the second anode; 3) to reduce the radius of the exit aperture of the second anode; 4) to reduce the beam current.

In reality, the first two measures do not lead to the desired result. Increasing the beam convergence angle leads to an increase in the beam cross-section in the region of the focusing field of the second lens. As a result, the aberration increases, as does the defocusing of the beam during deflection.

When the anode voltage is increased, as the results of experiments\(^{38,6}\) show, the spot size decreases only until the voltage reaches a certain, relatively small value, and then remains unchanged. A reliable explanation of this phenomenon has not yet been found.

Reducing the diameter of the aperture of the second anode and reducing the beam current make it possible to reduce the spot diameter substantially. The apertures are made so small that the manufacture of the projector parts and the subsequent assembly require exceptional precision. When magnetic focusing is used, very accurate positioning of the focusing coil is necessary.

In this way, M. Ardenne\(^{18}\), as early as 1939, obtained a spot with a diameter on the order of \(10\ \mu\).

It should be noted that when the beam current is reduced, higher current densities become attainable. Indeed, when the beam cross-section is reduced by means of a diaphragm, the space charge in the transverse section of the beam decreases and, consequently, the mutual repulsion of the electrons decreases.

The previous electron-repulsion forces are obtained already at a higher current density; consequently, focusing the beam after diaphragming will increase the current density in the beam. The mutual repulsion of electrons also leads to the fact that the smallest beam cross-section is obtained not in the focus of the projector, but somewhat farther away. This must be taken into account in designing the tube and, in particular, the projector.

The influence of the spread of initial electron velocities. Electrons leave the cathode in various directions, having a Maxwellian velocity distribution.

This spread of the initial velocities of the electrons in magnitude and direction weakens the focusing action of the projector, lowering the limiting attainable current density in the beam and also hindering reduction of the spot size.

Analysis shows¹ that, in order to weaken the influence of the spread of initial velocities, it is necessary: 1) to choose a cathode possessing a high emission current density at as low an operating temperature as possible*), 2) to increase the potential of the region in which the spot is formed, 3) to increase the convergence angle of the beam in the spot.

The first requirement is most fully satisfied by an oxide cathode. Satisfying the second requirement leads to difficulties with beam deflection. Concerning the possibility of fulfilling the third requirement, it has already been stated above in the discussion of the phenomenon of mutual repulsion of electrons.

*) Indeed, the lower the cathode temperature, the smaller the initial velocities of the electrons and the weaker the effect of the velocity spread will be.

In addition to the main factors considered, the spot diameter is also affected by such factors as the phenomenon of total internal reflection in the screen glass and the focusing action of the field of the post-acceleration electrodes. Both effects will be considered below.

§ 8. WRITING SPEED

When oscillographing rapidly varying processes, the speed of motion of the beam over the screen attains large values—\(10^9 \div 10^{10}\) cm/sec and even more. As the speed of motion of the beam increases, the brightness of the image decreases, and it may become so faint that it cannot be recorded. In the study of repetitive processes, repeatedly swept on the screen, the brightness may be sufficient if the phosphor has a long afterglow time. In this case the brightness of the luminescence does not have time to fall appreciably during the repetition period of the sweep voltage, and its accumulation over a number of periods takes place. A comparison of the phosphors most widely used in oscillography1 shows that for most of them the afterglow time lies within \(10^{-2} \div 10^{-3}\) sec, which is quite sufficient for oscillographing rapidly varying processes.

The investigation of single-shot processes is considerably more difficult because of the low brightness of the luminescence. It is evident that there exists a limiting speed of motion of the beam over the screen at which the image can still be distinguished. The value of this limiting speed depends on the method of recording and will be different for visual observation and for photography. Therefore, for high-speed oscillographic tubes a new, very important parameter is introduced—the maximum writing speed of the tube. By the maximum writing speed is meant the greatest speed of motion of the beam over the screen at which a sufficiently sharp image is obtained in a photograph. It is clear that data on the maximum writing speed will be incomplete unless the conditions of photography are specified (the aperture ratio of the lens and the sensitivity of the negative material).

The maximum writing speed can be determined by applying a harmonic oscillation2 to the deflecting system of the tube. On the screen an image is obtained which varies according to the law:

\[ y = A \sin \omega t . \]

The speed of motion of the beam over the screen is made up of two velocities: \(v_t\)—along the time axis, and \(v_y\)—vertically. The velocity \(v_t\) is constant in time, while \(v_y\) has its greatest value at zero deflection of the beam, equal to

\[ v_{y(y=0)} = \dot{y}_{(y=0)} = \omega A = 2 \pi f A . \]

The total speed of motion of the beam over the screen is

\[ v = \sqrt{v_t^2 + v_y^2} = \sqrt{v_t^2 + (2\pi f A)^2}. \]

By selecting the frequency or the amplitude of the oscillation, one can in this way determine the maximum writing speed. Here it is necessary to note the interrelation between the maximum writing speed and the resolving power of the tube. Let us compare two tubes with the same maximum writing speed, but with different deflection sensitivity at the same spot diameter. When the same voltage is applied to the deflecting systems of these tubes, in the tube with the lower sensitivity the speed

of the beam’s motion across the screen will be smaller. Consequently, the brightness of the image in this tube will be greater, and with its aid it will be possible to record faster processes. However, the resolving power proves to be lower than in the second tube. It is therefore necessary to reduce the spot diameter in order to preserve sufficient resolving power.

Among the factors on which the maximum recording speed depends, it is necessary to note the properties of the phosphor, the beam-current density, the accelerating voltage, the shape of the bulb, the sensitivity of the photographic emulsion, and the aperture ratio of the objective.

The characteristics of phosphors are described in detail in the literature34[^6][^41][^42]; therefore, in the present review it seems possible to characterize only those of their properties that are of special importance in the operation of high-speed oscillographic tubes. These properties of phosphors will be considered in connection with other factors affecting the brightness of the phosphor luminescence.

§ 9. DEPENDENCE OF BRIGHTNESS ON CURRENT DENSITY AND ACCELERATING VOLTAGE

The brightness depends most strongly on the beam-current density and the accelerating voltage.

It follows from experiment that the brightness of luminescence should increase in proportion to the increase in current density. However, at large current densities there occurs a saturation of brightness, which is determined mainly by the phosphor and by the method of its deposition.

On the other hand, this relation is fulfilled only up to a certain, quite definite value of the accelerating voltage, at which a limitation of brightness occurs, caused by the accumulation of negative charge on the phosphor.

Fig. 14. Dependence of the secondary-emission coefficient on the accelerating voltage.

Fig. 15. Dependence of the screen potential on the accelerating voltage.

In Fig. 14 the dependence of the coefficient of secondary emission of the phosphor on the accelerating voltage is presented. When the accelerating voltage is changed from \(U_A\) to \(U_B\), the coefficient of secondary emission is greater than unity, and the screen acquires a potential somewhat greater than the potential of the anode. For accelerating voltages greater than \(U_B\), the coefficient of secondary emission is less than unity; as a result, the screen is charged with a negative charge, which retards the electrons of the beam. The screen potential now remains unchanged, despite the increase in accelerating voltage (Fig. 15), and, consequently, the increase in the brightness of luminescence ceases.

This negative effect can be eliminated by means of a thin aluminum film applied to the side of the phosphor facing the projector345. A large positive potential is applied to the film, so that, owing to the good conductivity of aluminum, the beam electrons are easily removed from the screen. Experiments show that with the use of an aluminum coating the brightness increases considerably. Figure 165 presents the dependence of brightness on accelerating voltage for ordinary and aluminized screens. From these characteristics it is evident that the use of an aluminum coating is advantageous at accelerating voltages above 4 kV; moreover, for an accelerating voltage of 7 kV, the gain in brightness, compared with an ordinary screen, is twofold.

Fig. 16. Dependence of the brightness of the glow on the accelerating voltage for ordinary and aluminized screens. a) non-aluminized screen; b) screen coated with a thick layer of aluminum; c) screen coated with a thin layer of aluminum.

In addition, the aluminum film reflects toward the observer those rays from the luminous spot which go into the interior of the bulb. Of course, part of the reflected light energy is lost in the screen material, but nevertheless the brightness of the image increases somewhat.

Thus, by using a metallized screen, the brightness can be increased considerably by increasing the accelerating voltage. However, this leads to a large loss in sensitivity, since the flight time of the electrons in the deflecting system is reduced. The use of the post-acceleration principle makes it possible to increase the brightness considerably with only a small decrease in sensitivity678910. In this case the electron beam is first accelerated by a relatively small voltage of the second anode of the projector. After passing through the deflecting system, the beam electrons are accelerated by the post-acceleration electrode, whose potential is considerably higher than the potential of the second anode. The post-acceleration electrode is formed by an annular strip of conducting material applied to the inner surface of the bulb. A schematic representation of such a tube is shown in Fig. 17. Often not one but several post-acceleration rings are used. The graphite rings are separated by rings of nonconducting material. The voltage is supplied to the post-acceleration rings through the wall of the bulb.

Fig. 17. Usual form of a tube with post-acceleration: 1—ring of insulating material, 2—screen, 3—graphite coating connected to the second anode; 4—post-acceleration ring.

Table III

Since the energy of the beam electrons is determined by the final accelerating voltage, comparatively small voltages can be applied to the second anode, as a result of which the time of flight of the electrons through the deflecting system is reduced. Consequently, when acceleration after beam deflection is used, not only is the brightness increased, but the deflection sensitivity is also preserved.

As can be seen from Table III^47, a tube without post-acceleration, with an ordinary voltage, has low brightness at the required sensitivity. The same tube with an increased accelerating voltage has 35 times greater brightness, but with a considerable reduction in sensitivity. The advantages of these two tubes are combined in a tube with post-acceleration.

It should be noted that the sensitivity of a tube with post-acceleration is nevertheless somewhat lower than that of an ordinary tube. This is explained by the collecting action of the lens formed by the fields of the second anode and the post-acceleration electrodes. However, the decrease in sensitivity does not yet mean a deterioration in the resolving power of the tube. As the post-accelerating voltage is increased, the spot diameter decreases more rapidly than the sensitivity falls, and, consequently, the relative sensitivity increases.

To reduce the distortions introduced by the field of the post-acceleration electrodes, it is necessary to strive to make the equipotentials of this field perpendicular to the trajectories of the beam electrons. Some approximation to this ideal case is provided by the use of several post-acceleration rings, since in this case the accelerating potential along the trajectory of the electrons increases more smoothly than in a tube with one post-acceleration ring. As a result, in a tube with several stages of post-acceleration the ratio of the maximum accelerating voltage to the voltage of the second anode can be brought up to 10, whereas in a tube with one stage of post-acceleration this ratio does not exceed 3.

Thus, the introduction of post-acceleration makes it possible to increase the brightness significantly, and hence also the maximum recording speed.

§ 10. CONTRAST

The absolute value of the luminous brightness does not fully characterize the capabilities of a tube, since the recording device can register only such an image whose brightness is greater than the brightness of the other areas of the screen. Therefore another characteristic of the screen is introduced—“contrast.” By contrast is understood the ratio of the brightnesses of the brightest and the darkest areas of the screen.

Strongly reducing the contrast is the presence of external illumination of the screen. A decrease in contrast also occurs because of the phenomenon of total internal reflection in the screen glass (Fig. 18). Part of the rays from the luminous spot is reflected from the boundary between the glass and the air, and then, after secondary reflection, creates a halo around the spot and illumination of the remaining parts of the screen. In article^49 methods are presented for weakening this effect.

Fig. 18. Formation of a halo.

Some influence on the contrast is also exerted by the shape of the bulb. Rays from the luminous spot that go inside the tube may, after reflection from the walls of the bulb, reach the screen and illuminate it (Fig. 19, a). Therefore the bulb of the tube is given a special shape (Fig. 19, b), as a result of which the light rays return to the screen only after multiple reflections. In tubes with an aluminized screen, the light rays do not pass into the bulb; therefore, from this point of view, the shape of the bulb is immaterial.

Fig. 19. Influence of the shape of the bulb on the reflection of light rays falling from the screen.

§ 11. PHOTOGRAPHY

The final stage of most measurements on an oscillograph is photography. The photographic material may be placed either outside the tube or inside it. In the first case, the phosphor is excited by the electrons; after passing through the glass of the screen and the optical system, its glow acts on the photographic emulsion. However, only a small part of the electron energy is converted into light energy. In turn, only a small part of the light energy reaches the photographic emulsion. In the second case, with internal photography, almost all the electron energy is used. A substantial inconvenience of this method, however, is the need for continuous evacuation of the tube, since when the photographic material is changed air is admitted into the tube.

When photographing from the screen, it is necessary that the spectral regions of maximum light output of the phosphor and maximum sensitivity of the photographic emulsion coincide. Phosphors specially intended for photography give blue, violet (ZnS·Ag, CaWO₄), and whitish-blue (ZnS·CuZn, pure ZnS) colors⁴². Most of these phosphors belong to the sulfides and have their emission maximum in the blue part of the spectrum; however, a considerable part of the afterglow energy is emitted in the yellow and green parts of the spectrum. Therefore, for photographing from such screens it is advisable to use orthochromatic and panchromatic emulsions whose sensitivity maximum lies in the yellow-green part of the spectrum.

A number of questions concerning the technique of photography are covered in works⁵⁰–⁵⁵.

CONCLUSIONS

In conclusion, we shall dwell on the results achieved and on the prospects for the development of oscillographic tubes for fast processes.

At present the bandwidth of the investigated frequencies has been brought up to 10,000 Mc/s. Tubes with such a bandwidth make it possible to investigate processes with durations of the order of \(10^{-9}\) sec.

Most tubes possess good resolving power. The minimum value of the signal that can be resolved on the screen of the tube is of the order of a volt. An exception is the micro-oscillograph, for which this value amounts to tenths of a volt.

It may be considered that the maximum writing speed of the order of \(10^{10}\) cm/sec, obtained for a number of tubes, is a value actually attainable in industrial types of tubes. At such writing speeds the minimum resolving time reaches \(5 \cdot 10^{-11}\) sec.

However, frequency distortions in the deflecting system still do not allow

are able, with sufficient accuracy, to investigate those transient processes whose spectrum components exceed 1000 MHz. A large fraction of these distortions is due to insufficient matching between the signal transmission line and the deflecting system.

At the present time, oscillographic tubes are needed that make it possible to study transient processes with durations of \(10^{-9} \div 5 \cdot 10^{-11}\) sec and amplitudes of tenths of a volt. Consequently, in developing new tubes it is necessary to strive to improve the frequency characteristic of the deflection and the resolving power of oscillographic tubes.

References

1. M. Ya. Mulyarov, Electron-beam devices, Gosenergoizdat, 1954.
2. I. S. Stekolnikov, Pulse oscillography and its application, Academy of Sciences of the USSR, 1949.
3. I. S. Stekolnikov, Electronic oscillography of short-time processes, GTII, 1952.
4. I. S. Stekolnikov, The electronic oscillograph, Gosenergoizdat, 1949.
5. Electronic tubes, Part II, Sovetskoe Radio Publishing House.
6. Electron-beam tubes and indicators, Parts I and II. Sovetskoe Radio Publishing House, 1949—1950.
7. L. A. Meerovich, L. G. Zelichenko, Pulse engineering, Sovetskoe Radio Publishing House, 1953.
8. A. Rusterholtz, Electron optics, IL, 1952.
9. D. R. Hardy, The Recording of High-Speed Single Stroke Electrical Transients, Electronic Engineering, January, February (1956).
10. B. Jackson, The Recording of High-Speed Electrical Transient Phenomena. M. Sc. Thesis, Manchester University, 1953.
11. H. G. Rudenberg, Deflection Sensitivity of Parallel-Wire Lines in Cathode-Ray Oscillographs, J. Appl. Phys. 16, 279 (1945).
12. R. Feinberg, R. Jackson, D. R. Hardy, Design Features in Cathode-Ray Tubes for the Recording of High-Speed Transients, Nature 172, 1056—7 (1953).
13. L. R. Bloom, H. M. von Foerster, Ultra-High Frequency Beam Analyzer, Rev. Sci. Instr. 25, 649—653 (1954).
14. R. C. Hergenrother, H. G. Rudenberg, A Direct-Reading Oscilloscope-Frequenzen, Z. Tech. Phys. 18, 256, 1937.
15. D. M. Mackay, Some New Aspects of High-Speed Oscillography. Proc. Phys. Soc. 61, 235, 1948.
16. A. M. Chernushenko, An oscillograph for the investigation of UHF oscillations, Radio Engineering and Electronics, Vol. I, No. 3, Academy of Sciences of the USSR Publishing House, 1956.
17. M. von Ardenne, Der Electronen Mikroszillograph Hochfrequenztechnik und Elektroakustik 54, 181 (1939).
18. M. von Ardenne, A Six-Trace Cathode-Ray Micro-Oscillograph, Wireless. Engin. 19, 231 (1942).
19. M. von Ardenne, Ein Präzisions-Elektronenstrahloszillograph mit Mengen u. Schreibfleckdurchmesser Nachrichtentechnik 5, 481—492 (1955).
20. G. M. Lee, A Three-Beam Oscillograph for Recording of Frequencies up 10000 Megacycles, Proc. Inst. Radio Engin. 34, 121 (1946).
21. J. Fert, J. Lagasse, J. Olle, Sur la Realisation d’un oscillographe cathodique à plusieurs faisceaux, C. R. Acad. Sci. Paris 59, 238 (1954).
22. J. R. Pierce, Travelling Wave Oscilloscope, Electronics 22, 197 (1949).
23. R. Owaki, S. Terahata, T. Hada, T. Nakamura, The Travelling Wave Cathode-Ray Tube, Proc. Inst. Radio Engrs. 38, 1172 (1950).
24. S. T. Smith, R. V. Talbot, C. H. Smith, Cathode-Ray Tube for Recording High-Speed Transients, Proc. Inst. Radio Engrs. 40, 297 (1952).
25. S. G. Konstantinov, Diploma project. Supervisors G. M. Topchiev and Yu. M. Shamaev. Molotov Moscow Power Engineering Institute, 1953.
26. A. M. Chernushenko, Communication at the IRE, Academy of Sciences of the USSR. Radio Council of the Academy of Sciences of the USSR.
27. Yu. M. Shamaev. New methods of oscillographing superfast processes. In Assistance to Industry. Collection of annotations of scientific-research works, issue 5. Molotov Moscow Power Engineering Institute Publishing House, 1952.
28. Yu. M. Shamaev, Dissertation, Molotov Moscow Power Engineering Institute. 1952.

1. H. E. Hollman, The Travelling Wave Cathode-Ray Tube, Proc. Inst. Radio Engrs. 19, 194 (1951).

2. E. W. Ernst, H. von Foerster, Electron Bunches of Short Time Duration. J. Appl. Phys. 25, 674—675 (1954).

3. Paul D. Coleman, M. D. Sirkis, Crossed Electron Beam Technique for Displaying Short Pulses. J. Appl. Phys. 24, 952 (1953).

4. Paul D. Coleman, D. Murray, D. Sirkis, Irving Kaufman, Crossed Electron Beam Technique for Measuring Space—Charge Effects in Beams. J. Appl. Phys. 25, 679 (1954).

5. D. B. Langmuir, Theoretical Limitations of Cathode-Ray Tubes, Proc. Inst. Radio Engrs. 25, 977 (1937).

6. H. Moss, The Electron Gun of the Cathode-Ray Tube, Part I, Journ. of the Brit JRE 5, 10 (1945).

7. J. R. Pierce, Theory and Design of Electron Beams. New York, 1949.

8. J. G. Maloff and D. W. Epstein, Electron Optics in Television. Mc Grow-Hill Book Company, 1938.

9. E. Gundert, Der Öffnungsfehler von elektrostatischen Rohreinsen, Zeits. Phys. 112, 689 (1939).

10. J. A. D. Lewis, F. H. Wells, Millimicroseconde Pulse Technique N Y ÷ L, 1954.

11. A. V. Moskvin, Cathodoluminescence, Part I, Gostekhizdat, 1948.

12. G. F. J. Garlick, Cathodoluminescence, Advances in Electronics 2, 151—184. Academic Press Ins. Publishers, New York, 1950.

13. G. F. J. Garlick, The Physics of Cathode-Ray Tube Screens Electronic Engineering. 287, August, 1949.

14. N. F. Zhirov, Luminophores. State Publishing House of the Defense Industry, 1940.

15. R. W. Duddling, Aluminium Backed Screens for Cathode-Ray Tubes. Journ. of the Brit. JRE 11, 455—462 (1951).

16. E. Schwartz, Am Stande des Nachbeschleunigungsprobleme bei Kathodenstrahlröhren Fernseh. A. G., t. I, pp. 19, 1938.

17. J. A. Gier, A Cathode-Ray Tube with Post-Acceleration, Philips Techn. Rev. No. 9, 245—252 (1940).

18. J. R. Pierce, After-Acceleration and Deflexion, Proc. Inst. Radio Engrs. 29, 28 (1941).

19. W. G. White, Cathode-Ray Tube with Post-Deflexion Accelaration, Electronic Engineering 21, 75 (1949).

20. L. S. Allard, An «Ideal» Post Deflexion Accelerator C. R. T., Electronic Engineering 22, 461 (1950).

21. A. E. Martin and R. M. Bowie, Picture-Tube Contrast Improvement, Electronics 110, August, 1950.

22. R. G. Hopkinson, The Photography of Cathode-Ray Tube Traces. Journ. Inst of Electrical Engrs. 93, Pt3A, p. 808 (1946).

23. R. J. Hercock, R. G. Hopkinson, W. F. Berg, W. Nethercot, The Photographic Recording Cathode-Ray Tube Screen Traces, Photogr. Journ. 86B, 138 (1946).

24. R. J. Hercock, The Choice of Emulsions and Developers, Photogr. Journ., 86B, 138 (1946).

25. R. J. Hercock, The Recording of Cathode Ray Tube Traces. Illford Ltd, p. 11, 1947.

26. W. F. Berg. Photographic Aspects of High-Speed Recording Photogr. Journ. 86B, 154, 1946.

27. G. S. Moore, The Last Ounce of Speed, Photogr. Journ. 88A, 239 (1948).

3. Reference 39. ↩↩

4. Reference 40. ↩↩

5. Reference 43. ↩↩

6. Reference 44. ↩

7. Reference 45. ↩

8. Reference 47. ↩

9. Reference 48. ↩

10. Reference 49. ↩