Cathode-Ray Oscillograph*

J. B. Johnson, New York

There are several different types of oscillographs. The type to which this article is devoted uses a stream of cathode rays and is therefore called a cathode-ray oscillograph. The principle of its operation is very simple. In this oscillograph there are two electrodes in a long evacuated glass tube (Fig. 1); one of these electrodes may be an incandescent filament, the other is a plate with a small aperture. If a potential difference is applied between the electrodes, so that the filament is the cathode and the plate the anode, then the electrons emitted by the incandescent filament are drawn toward the anode. Some of them pass through the narrow aperture in the anode and travel in the form of a thin beam of electrons, in the form of cathode rays, along the tube. At the end of the tube is placed a screen made of a fluorescent substance, which glows brightly at the point where the cathode rays fall upon it. In this way we can see the place where the cathode rays strike the screen. In addition, another pair of electrodes is placed in the tube in the form of two plates

Fig. 1. Diagram of a cathode tube.

* J. B. Johnson, The Bell System “Technical Journal,” 11, 1, 1932.

\(P\) and \(P_1\), so that the cathode rays pass between these plates (Fig. 2). By means of a battery, or in some other way, we apply a potential between these plates, so that one of them will be positive with respect to the other. Since the electrons of the cathode ray are negative charges, as they pass between the plates,

Fig. 2. Tube with electrostatic deflection.

they are attracted to the positive plate and emerge having changed their direction under the influence of the applied potential. In the same way, a magnetic field \(NS\) (Fig. 3), situated perpendicular to the path of the cathode ray and

Fig. 3. Tube with magnetic deflection.

in the plane of the drawing, causes the cathode ray, after passing through the field, to deviate out of the plane of the drawing. The magnitude of the deflection serves as a measure of the intensity of the applied magnetic or electric field. Thus, in this cathode ray we have an indicator that shows us the magnitude of the field deflecting it. But this pointer is remarkable in that it has almost no mass and iner-

Cathode Oscillograph

...ideas. Therefore it can follow changes in the applied field extremely rapidly. In view of this feature, this instrument is very widely used in the study of electrical phenomena of the most varied kinds, as in electric machines, telephone apparatus, radio transmitters, and electric waves. One of the most highly effective applications of the cathode oscillograph is connected with the study of lightning; this is, in all probability, the most interesting work devoted to lightning since its electrical nature was discovered by Franklin 180 years ago.

Let us consider more closely the action of this tube. The velocity of the electrons as they emerge from the aperture in the anode can be determined from the energy equation,

\[ \frac{1}{2}mv^2 = eV, \]

where \(V\) is the potential between the cathode and the anode, \(e\) is the charge of the electron, and \(m\) and \(v\) are its mass and velocity. From this equation we obtain the following relation between the velocity and the accelerating potential:

\[ v = \sqrt{2\frac{e}{m}V}. \]

The quantity \(\frac{e}{m}\), as is known, is equal to \(1.77 \cdot 10^7\) CGSM; 1 volt amounts to \(10^8\) CGSM, and therefore the velocity of the electrons is expressed as

\[ v = 5.95 \cdot 10^7 \sqrt{V}\ \text{cm/sec}. \]

If the accelerating potential is \(300\ \mathrm{V}\), then the velocity of the electrons according to this formula, in round numbers, will be \(20 \cdot 10^9\ \text{cm/sec}\). In a tube \(20\ \text{cm}\) long, the electron travels from the deflecting plates to the screen in \(20 \cdot 10^9 = 1/50\,000\,000\ \text{sec} = 5 \cdot 10^{-8}\ \text{sec}\). If the applied potential is \(30\,000\ \mathrm{V}\), then the velocity is approximately 10 times greater than at \(300\ \mathrm{V}\); it will therefore be equal to \(1/3\) the speed of light. The change in the direction of the ray, imparted to it by the deflecting plates, is therefore transmitted to the end...

of the beam in a very short interval of time, and the beam faithfully transmits very rapid oscillations of the potentials of the plates.

Let us now consider how the beam reacts to a potential applied to the plates. Let the beam propagate normally along the tube with velocity \(v\), and let (Fig. 4) it pass between two plates of length \(l\), separated by a distance \(d\) and having a potential difference \(V'\). While the beam passes between the plates, the electrons experience an acceleration

Fig. 4. Electrostatic deflection.

\[ a=\frac{e}{m}E=\frac{e}{m}\cdot\frac{V'}{d}. \]

This continues for a time

\[ t=\frac{l}{v}. \]

Therefore the velocity in the perpendicular direction acquired during this time will be:

\[ v'=at=\frac{e}{m}\frac{V'}{d}\cdot\frac{l}{v}. \]

After this the beam travels in a straight line to the screen, which it intersects at a distance \(D\) from the normal position. The deflection \(D\) is determined from the following obvious relation:

\[ \frac{D}{L}=\frac{v'}{v} =\frac{e}{m}\cdot\frac{V'}{d}\cdot\frac{l}{2\frac{e}{m}V} =\frac{1}{2}\cdot\frac{l}{d}\cdot\frac{V'}{V}, \]

\[ D=\frac{1}{2}\cdot\frac{lL}{d}\cdot\frac{V'}{V}. \]

The expression obtained gives the following interesting design indication: for high sensitivity the plates must be long and located close to one another, but they must not intersect the path of the deflected beam. If we wish to obtain some maximum deflection \(D\) with a tube of a definite length \(L\), then the ratio between the distance between the plates and their length must be as follows:

\[ \frac{d}{l}=\frac{D}{L}, \]

as is easily seen from Fig. 4.

Fig. 5. Magnetic deflection.

The magnetic sensitivity of the tube is more indefinite, because the boundaries of the magnetic field are less sharp than the boundaries of the electric field. If the beam, produced by the accelerating potential \(V\), passes through a region of width \(l\) (Fig. 5) in which there is a transverse magnetic field of strength \(H\), then the path of the beam will be an arc of a circle of radius

\[ R=\frac{mv}{iH}=\frac{1}{H}\sqrt{\frac{2mV}{i}}. \]

On leaving the magnetic field, the beam continues its path in a straight line to the screen, where let its total deflection from the normal be \(D\). If the angle of deflection \(\theta\) is small, then, with a good approximation, we have:

\[ \operatorname{tg}\theta=\frac{D}{L}=\frac{l}{R}=lH\sqrt{\frac{i}{2mV}}, \]

whence

\[ D = iLH \sqrt{\frac{l}{2mV}} . \]

If practical units are introduced instead of electromagnetic units, the expression takes the form:

\[ D = \frac{3lLH}{V \sqrt{V}} . \]

Fig. 6. The first cathode oscillograph, F. Braun, 1897.

After this elementary exposition of the operation of the cathode oscillograph, let us turn to the history of the development of this instrument.

The first indication of the idea that cathode rays could be used for measuring a magnetic field dates to 1894, when Hess² in France proposed using such a tube for tracing curves. However, the first application of this idea belongs to Ferdinand Braun (1897)³; since then the instrument has been called the Braun tube.

Fig. 7. Tube for measurement, J. J. Thomson, 1897.

The Braun tube has a very simple construction (Fig. 6).

The cathode in it is a disk, the anode is a wire soldered on the side; further, there is a diaphragm in the form of a drilled hole and a fluorescent screen of zinc sulfide. There is air in the tube at reduced pressure. The current from an electrostatic machine creates a discharge in the remnants of the gas, as a result of which cathode rays appear.

It is interesting to note that the invention of the tube was made before the nature of cathode rays had been deciphered. This happened in the very year when J. J. Thomson in England and Kaufmann in Germany, each using tubes almost identical with Braun’s tube, showed that cathode rays possess charge and mass. Thomson’s tube is shown in Fig. 7.

Braun’s tube immediately found various applications.

Fig. 8. Ebert and Hoffmann, 1898 — “Geissler tube.”

One of the most fruitful areas for such application proved to be the field of radio transmission. Numerous works in this direction were carried out by Zenneck and his school.

After the invention of this tube, various improvements were introduced into it, which made its use more convenient. Figs. 8–12 show some tube designs of that time. In 1905 Wehnelt proposed using a hot cathode coated with oxide

some alkaline-earth metal; such cathodes had, several years earlier, proved to be convenient sources of intense electron radiation. Wehnelt designed

Fig. 9. Mack Gregor—Morris, 1902 — Kossor tube⁶.

a tube (Fig. 13) that could operate on the mains at 220 V. This was probably the first application of oxide cathodes. In the following years a considerable number of experimental tubes with heated cathodes were built,

Fig. 10. Ryan, 1903 — Müller—Uri tube⁷.

but almost 20 years passed before a truly successful tube with a heated cathode, suitable for operation on a low-voltage mains supply, was designed. This was

Fig. 11. Rozhanskii, 1911⁸.

the Western Electric tube No. 224¹¹, which we shall describe in detail below.

Another tube of this type, built by Arden and Gartel¹², is shown in Fig. 14.

After Braun and Wehnelt, the most substantial change in the design was made by Dufour in France

Fig. 12. Brereton, 1913 — Max Kohl tube².

Fig. 13. Venen, 1905.

Fig. 14. Arden-Hartel, 1930 — Leibold tube.

in 1914. Until that time, the image obtained on the fluorescent screen was recorded by means of a photographic camera. This usually required repeated reproduction of the image occurring in a rapid process in order to obtain sufficient exposure.

Fig. 15. Dufour oscillograph, 1923¹⁴.

Fig. 16. Wood, 1923¹⁵.

Dufour lowered the fluorescent screen and placed a photographic plate inside the tube in such a way that the cathode rays could act directly on the plate. If the cathode rays fall directly on the photographic emulsion, the recording can be made in a considerably shorter interval of time than in that

case, when the light of the fluorescent screen is focused by means of a lens onto a photographic plate. The introduction of the photographic plate into the instrument, of course, entailed a certain complication of the design: there arose the need for a mechanism to move the plate inside the tube for removing plates and putting them in place; there also arose the need for pumps to obtain and maintain the vacuum. The former glass structures were almost completely abandoned and replaced by metal ones. Taken all together, this assumed the form of complicated apparatus, which, however, proved very useful.

Fig. 15. Westinghouse-Norinder, 1928.

In the last few years several designs of tubes of this type have been developed. Along with Dufour’s tube one may name Rogowski’s tube in Germany, Wood’s in England, Berger’s in Switzerland, Norinder’s in Sweden, and those of the Westinghouse Company and the General Electric Company in America (Figs. 16–18, 21). In all these instruments we had a complicated construction of the tube and of the control grid. Some tubes were built so that they operated during only one impulse of a wave of 60 periods. In others

there was an ingenious switching device, which turned on the tube at the very beginning of the electrical pulse being studied, and the tube then operated, recording the rest of the pulse. If we take into account that im-

Fig. 18. Norrinder, 1930—the oscillograph tube. ^18

pulses can arise as a result of a lightning strike on a line, then we shall understand that there may exist something “faster than lightning.”

In recent years a further step forward has been made by Max Knoll ^16. He considerably simplified the operation of the tube by equipping its end with a thin window similar to a cathode-

tube. Thus the cathode rays, having passed through a thin screen of metal or cellophane, fell upon a photographic plate located outside the tube.

Thus we have three types of cathode oscillographs: 1) an oscillograph of the Braun-tube type, consisting of a glass tube with a fluorescent screen and requiring a relatively high voltage; 2) tubes of the type of the Dufour tube just described, with direct recording on a photographic plate or film; 3) tubes with a hot cathode and with a relatively low operating potential.

The Western Electric No. 224 cathode oscillograph belongs to this last type. Since this is precisely the tube with which I have worked directly, I shall dwell on certain problems connected with its construction and operation.

The diagram of this tube, shown in Fig. 19, is now widely known*, and I shall describe it only insofar as it is necessary to discuss the rationale for one or another of its features.

First of all, we need a convenient tube for work on a ba—

Fig. 19. Western Electric tube.

* After this report was read, the tube was modified in certain respects. The glass envelope for the filament was replaced by a metal one; the anode and the deflecting plates were mounted on stamped insulated supports. These changes make the construction more stable and ensure, to a greater degree, accurate positioning of the various parts. The end of the tube was modified in such a way that its surface was made cylindrical instead of spherical, which permits more direct contact between the fluorescent screen and the photographic film.

tures of ordinary vacuum tubes. Therefore a cathode in the form of a heated filament is necessary. The anode is a metal tube placed at a short distance from the cathode, with a metal disk with holes situated between them, through which the electrons pass to the anode. These electrodes are shown in Fig. 20 and are designated, respectively, by the letters \(C\), \(A\).

The electrons go from the cathode into the anode, some of them passing through the anode and forming an electron beam. Further, for reasons that will be indicated below, there is a certain amount of gas in the tube, and this

Fig. 20. Diagram of an electron gun.

imposes two requirements on the construction of the electron gun. First, the presence of gas produces considerable ionization inside the tube when electrons enter it. As a result, a large part of the current would be directed to the disk and would pass outside the anode. Therefore the cathode and anode are enclosed in a small tube whose dimensions are less than the electron mean free path in the gas, so that no appreciable ionization can arise. However, some ionization is produced in the space between the cathode and the anode, and the positive ions are directed toward the filament. If the filament is not protected, the oxide layer is destroyed under the action of this bombardment in two or three hours. Therefore the filament is wound in the form of a spiral, located coaxially with the anode and with the hole in the disk, so that it lies outside the path of the ions. In this way the filament is preserved for several hundred hours of operation.

This exhausts the internal parts of the electron gun. On the outside, two pairs of deflecting--

...ing plates, which control the motion of the electrons after they have flown out of the gun. In order to avoid large potential differences between the anode and the deflecting plates, one plate of each pair is connected directly to the anode, and only the other acquires the alternating potential imparted to it. As

Fig. 21. General Electric oscilloscope, 1928.

for the dimensions and spacing of the plates, they are calculated so as to create maximum sensitivity for the given full deflection. The sensitivity is approximately 1 mm of deflection for each volt applied to the deflecting plates, or 1 mm for each ampere-turn in a pair of small coils located outside the tube. These figures are given for the normal accelerating potential of 300 V.

The flat end of the tube, where the electron beam strikes, is coated with a fluorescent substance. The fluorescing...

the fluorescent powder is a mixture of zinc silicate and calcium tungstate, specially prepared to obtain maximum luminescence. Zinc silicate gives a green glow, to which the eye is very sensitive, while calcium tungstate gives a violet glow, distinguished by high photographic activity, so that with this mixture one and the same tube can be used both for photographic and for visual observations.

As was said earlier, a certain amount of gas remains in the tube. The purpose of this gas is to create a small ionization in the tube so as to prevent the accumulation of excessively large charges on the glass walls and on the screen. Electrons settling on the glass are neutralized by positive ions produced in the gas. An electron current, equal to the current in the beam, passes in the opposite direction through the gas to the anode. The chief result of this movement of electrons is that a negative space charge is formed in the tube, which reduces the velocity of the electrons before they reach the screen, as though the accelerating potential had been reduced by approximately 50 V.

Another, more important purpose of the gas is to bring the electrons in the beam to a sharper focus on the screen. The beam is divergent for two reasons: first, because it is so from the very beginning, and secondly, as a result of the natural electrostatic repulsion of the individual electrons in the beam. The gas serves to compensate these effects in the following way: when the electron beam travels along the tube, some electrons collide with gas atoms and break these atoms up into an electron and a positive ion. The impact of an electron practically does not displace the massive positive ion from the position it occupied at the moment of impact, while the two electrons are immediately thrown out of the path. As a result, along the length of the beam there is formed a column of positive ionization with a negative space charge surrounding it. This creates a radial

CATHODE OSCILLOGRAPH

an electrostatic field which tends to bend the trajectory of the outer electrons of the beam inward, toward the center. The magnitude of this action depends on the degree of differential ionization. The latter, in turn, is the greater the higher the gas pressure and the stronger the current in the beam. The gas pressure must be sufficiently low so that the greater part of the electrons reaches the screen, and the current in the beam must be such as to produce the desired focusing action. The heavier the ions, the lower the pressure may be. Thus the conditions for focusing depend on the kind and pressure of the gas, the velocity of the electrons, the strength of the current in the beam, and the length of the tube. In tube 224, filled with argon at a pressure of 0.01 mm, focusing occurs at approximately a current of 20 μA in the beam. Smaller currents produce a large unfocused spot; larger currents produce a focus before reaching the screen, with a corresponding enlargement of the spot on the screen. Thus the spot on the fluorescent screen is focused by regulating the heating current of the cathode.

In addition to the focusing of electrons and the prevention of charge accumulation on the tube, the gas also plays a role in various other respects. A very curious effect of the gas consists in the fact that it lowers the sensitivity of the tube at small deflecting potentials. If a uniformly varying voltage is applied to a pair of deflecting plates so that the spot moves across the screen, then for an instant the spot seems to oscillate about the center of the screen and appears brighter there. This effect has always been observed, but special attention was recently given to it by Prof. Bedell[^20]. The explanation is connected with the space charge between the deflecting plates. The electron beam creates slowly moving positive ions and electrons in the gas along its path between the plates. When a voltage is applied to the plates, the positive ions are directed from the beam toward the negative plate, and an equal number of electrons are directed toward the positive plate. The space charge formed by the electrons and ions creates an electric field, opposit—

positive field caused by the applied potential. The greatest space charge is produced at the negative plate, where slowly moving positive ions accumulate. The space between the plates in the middle remains almost free of field, and here the deflection of the beam is insignificant until the voltage becomes greater than that at which all the ions formed will be carried to the plates. Calculation in agreement with experiment shows that this voltage is 2–3 V on either side of zero.

Fig. 22. Circuit for regulating the charge of a capacitor.^11

Now we shall finish the description of the construction and operation of certain oscillographs and proceed to consider their applications.

A cathode oscillograph is, in essence, an instrument that traces, in rectangular coordinates, curves giving the relation between two quantities represented by the fields between the deflecting plates. Often one of these quantities is time, as in the ordinary oscillograph with a moving mirror, while the other is some electrical quantity. In such a case we say that we are recording the waveform. For such recording we must be able to make the spot move with constant velocity. One of the simplest and most suitable methods for producing a linear time axis,^21 suitable at least for low-voltage tubes, is based on the use of two thermionic tubes for regulating the charge of a capacitor, as shown in Fig. 22. One tube here is a simple

two-electrode tube \(T_1\), limiting the charging of capacitor \(C\), so that the voltage on the capacitor increases linearly with time according to the equation

\[ V = Cit. \]

The second tube \(T_2\) is filled with gas and has the property of allowing current to pass only in the case when the voltage on it reaches a certain value, which, in turn, is regulated by the grid potential. When the potential of the capacitor reaches this value, the tube immediately makes it possible for the capacitor to discharge, and then the process of uniform charging begins again.

Fig. 23. Discharge of a capacitor through an inductive resistance.

The capacitor is connected to one pair of plates of the oscillograph, so that the spot, under the influence of the increasing potential of the capacitor, moves uniformly across the screen in one direction and then, with considerably greater speed, returns in the opposite—

Fig. 24. Discharge of a capacitor through a vibrating contact.

Fig. 25. Action currents of an excited frog nerve.^22

in the opposite direction. This process may be repeated once per second or many thousands of times per second, depending on the frequency of the wave being investigated.

Fig. 26. Closing of the current by direct contact and opening through a tuned oscillatory circuit,—Dufour.^14

In Figs. 23–29 records are shown that were made on a time scale. Figs. 23–25 are photographs taken with a “Western Electric” oscillograph; Figs. 26–27, with a Dufour oscillograph; and Figs. 28–29, with a Rogowski-type oscillograph.

Another way of using the oscillograph consists in recording relations independently of time. As a simple example one may point to the current–voltage curve—

... for a resistance through which an alternating current passes. The spot moves back and forth along a straight line, the slope of which characterizes the reciprocal value of the resistance (Fig. 30a). If a self-inductance is connected to the resistance, then the spot no longer moves along a single straight line; the resulting ellipse tells us the magnitude of the inductive resistance (Fig. 30b). This method opens up very varied possibilities of application. For example, instead of a resistance we may take a gas discharge whose properties we wish to study. One of the applications of this method is the recording of hysteresis curves of ferromagnetic materials ^25 (Fig. 31). Fig. 32 illustrates the application of this method to the study of distortion in an amplifier.

Fig. 27. Waveform at 8,500,000 cycles — Dufour, 1914.

Fig. 28. Front of a potential wave propagating in a conductor.

Let us suppose that we apply to each pair of deflecting plates a potential from two different oscillators. If the oscillators make exactly the same number of oscillations per second, the pattern in the tube remains in a stationary state; but if the frequencies of the oscillators differ, even by an insignificantly small amount, the pattern changes in accordance with the different phase relations of the two oscillators. This method is one of the most sensitive for comparing and calibrating precision oscillators; we may call it the method of Lissajous figures. Fig. 33 shows the form of some of these stationary Lissajous figures.

Fig. 29. Beginning of a spark in gas ^24.

Fig. 30. Diagram of the current–voltage curve.

Another method of comparing oscillator frequencies was called the “sawtooth” method. In this method the potential from a low-frequency source is split into two equal …

components differing in phase by 90°. These two potentials are applied to the deflecting plates, and a stationary circle is obtained on the screen. The potential from an oscillator of higher frequency is introduced into the circuit between the cathode and the anode of the tube, so that the sensitivity of the tube varies in accordance with the greater of the frequencies being compared. Therefore the circle is distorted into a toothed form, as shown in Fig. 34, under the condition that the higher frequency is an exact multiple of the lower. If the ratio of the frequencies is not a rational number, the toothed figure rotates, revealing a lack of synchronization.

Fig. 31. Hysteresis curve: a) iron, b) permalloy.

Fig. 32. Distortion in a cathode amplifier.1

Two interesting applications to radiotelephony are illustrated by Figs. 35 and 36. Fig. 35 gives two character-

2 : 1            5 : 4

8 : 1

Fig. 33. Frequency comparison.

istics of transatlantic communication; here the tube records the amplitude of the incoming signal as a function of the number of modulating frequencies. In Fig. 36 a diagram is given of the magnitude and direction of atmospheric disturbances as they are detected on the screen of the oscillograph tube.

Fig. 34. Frequency comparison by means of a “gear.”

For purposes of demonstration the tube may be used to determine the ratio of the electron charge to its mass, \(e/m\). The classical method is evident from the derivation of the sensitivity

...tube, given above. Namely, the quantity \(\frac{e}{m}\) can be obtained from two basic relations:

\[ \frac{1}{2}mv^{2}=eV;\quad mv=eHR. \]

This method is associated with certain errors, since the deflecting field cannot be determined exactly; moreover, errors are caused by the space charge of the tube.

Fig. 35. Transmission characteristic of transatlantic short-wave radio communication \(^{27}\).

More precise method of determining \(e/m\) belongs to H. Busch \(^{28}\). A long solenoid through which a direct current flows creates a uniform magnetic field in the tube parallel to its axis. The electrons move in this field along spirals in such a way that, when the field has one of a number of definite values, the electrons are focused on the screen. These critical values of the magnetic field are given by the equation

\[ H=\frac{2\pi}{L}\sqrt{\frac{2m}{e}\,V}; \]

where \(n = 1, 2, 3\), etc., \(L\) is the length of the bundle and \(V\) is the accelerating potential.

In this method smaller errors are obtained if the filament current is made so small that, after magnetic focusing, only a barely perceptible spot is produced.

Fig. 36. Azimuthal distribution of atmospheric disturbances.

From the very beginning the cathode oscillograph was recognized as an instrument promising great possibilities for application. However, its use was limited by the difficulties of maintaining a constant vacuum, of finding a suitable source of high potential, and also by the bulkiness of the apparatus. We have seen that in the almost 30 years that have passed

since the invention of the tube, very substantial improvements have been made both in the design of the tube and in the methods of its use. The introduction of tubes operating at low potentials, possessing high sensitivity and moderate cost, rapidly ensured the spread of the cathode-ray oscillograph, so that at the present time they are used in almost every laboratory where high-frequency measurements are made.

LITERATURE

1. A more detailed account of the history of the cathode-ray oscillograph is given by Rausurath (“Apparatus and Technique for Producing and Recording Curves of Alternating,” etc., Helios, 1912) and by MacGregor Morris and Mains (“Measurements in Electrical Engineering by Means of Cathode Rays”), Jl. Inst. El. Eng., 63, p. 1656, 1925.
2. Hess A., Compt. rend., 119, p. 57, 1894.
3. Braun F., Nied. Ann., 60, p. 552, 1897.
4. Thomson J. J., Phil. Mag., 44, p. 293, 1897.
5. Ebert u. Hoffmann, E. T. Z., 19, p. 405, 1898.
6. Mc. Gregor-Morris. Engineering, 73, p. 754, 1902.
7. Ryan. H. J., Am. Ins. E. Eng. Trans., 22, p. 539, 1903.
8. Roschansky D., Ann. d. Phys., 36, p. 281, 1911.
9. Broughton H. H., Electrician, 72, p. 171, 1913.
10. Wehnelt A., Phys. Zeit., 6, p. 732, 1905.
11. Johnson J. B., Jl. Am. Opt. Soc. R. S. I., 6, p. 701, 1922.
12. Hartel H. von, Zeit. f. Hochfr. Techn., 34, p. 227, 1929.
13. Dufour A., Comptes rend., 158, p. 1339, 1914.
14. Dufour A., Oscillographe cathodique, Etienne Chiron, Paris, 1923.
15. Wood A. B., Phys. Soc. Lond. Proc., 35–2, p. 109, 1923.
16. Knoll Max., Z. f. techn. Phys., 10, p. 28, 1929.
17. Norinder H., A. I. E. E. Trans., 47, p. 446, 1928.
18. Norinder H., Zeit. f. Phys., 63, p. 672, 1930.
19. Lee E. S., C. E. Rev., 31, p. 104, 1928.
20. Bedell F. u. Kuhn J., Phys. Rev., 36, p. 993, 1930.
21. Samuel A. L., Rev. Sci. Inst., 2, p. 532, 1932.
22. Gasser u. Erlanger, Am. Jl. Physiol., 73, p. 613, 1925.
23. Krug, W. E. Tz., 51, p. 605, 1930.
24. Krug, W. Zeits. f. techn. Phys., 11, p. 153, 1930.
25. Johnson J. B., Bell System Tech. Jl., 8, p. 286, 1929.
26. Willis u. Melhuish, Bell System Tech. Jl., 5, p. 573, 1926.
27. Potter R. K., Inst. Radio Eng., 18, p. 581, 1930.
28. Busch H., Phys. Zeit., 23, p. 438, 1922.
29. Watson, Watt u. Herd. Jl. I. E. E., 64, p. 611, 1926.

1. As visible in the source page. ↩