US2233878A - Electron multiplier - Google Patents

Description

March 1941- R. L. SNYDER ELECTRON MULTIPLIER Filed June 5, 1937 2 Sheets-Sheet 1 l w n o INVENTOR, RICHARD L. SNYDER A 'ITORNE YS.

March 4, 1941. I R SNYDER 2,233,878

ELECTRON MULTIPLIER Filed June 5, 1937 2 Sheets-Sheet 2 1N VEN TOR. RICHARD L. SNYDER.

ATTORNEYS.

Patented Mar. 4, 1941 UNITED STATES ELECTRON MULTIPLIER I I Richard L. Snyder, Glassboro, N. J., assignor, by mesne assignments, to Farnsworth Television 8; Radio Corporation, Dover, Del., a corporation of Delaware Application June 5, 1937, Serial No. 146,641

18 Claims.

My invention relates to electron multipliers operating by secondary emission, and more particularly to such an electron multiplier where ion bombardment of surfaces sensitized for secondary emission has been greatly reduced.

The present application, while embodying the same broad method as disclosed and claimed in the Philo T. Farnsworth application, Serial No. 80,193, filed May 16, 1936, now United States Patent No. 2,140,832 issued December 20, 1938, and entitled Means and method of controlling electron multipliers," differs from that prior application in that the method is practiced without additional electrodes positioned and energized to remove ions from the electron stream, and the method is applied herein to electron multipliers energized by unidirectional potentials.

In the above-identified Famsworth patent, it

' was pointed out that it was advantageous to re- 20 move ions from the electron paths during a multiplication process wherein electrons were multiplied by secondary emission caused by repeated impacts of an electron stream against an activated surface. Inasmuch as the majority of the best secondary emitting surfaces are usually composed of materials having relatively low melting points and relatively high vapor pressures, some ionization by collision takes place as the electrons traverse their normal paths between secondary producing impacts. Such ions as are produced are objectionable because they cause interference with thenormal multiplication cycle. By far the major result of such ionization, even though small in an electron multiplier, is that, if not controlled, the ions will bombard the sensitized surfaces and eventually spoil the surfaces for maximum secondary emission, and thus greatly reduce the life of the multiplier. If, then, the ions can be prevented in any manner from bombarding the sensitized surfaces, a much longer life is obtained, and in addition, uniformity in quantity production can be obtained, as well as uniformity of decadence curves.

It is therefore one object of my present invention to provide an electron multiplier wherein ions produced by collisionor otherwise along the electron paths in an electron multiplier may be separated from the paths, at least to the extent that they will impact an area of the electrodes which is not active for the production of secondary electrons by electron impact.

- The second major object of my invention is to provide an electron multiplier which may be built with an indefinite number of stages, and wherein the space currents and space charge densities may be kept approximately constant in the successive stages, so that the device will present a linear characteristic. A further object of my invention is to provide a multiplier of the type described in which the effectiveness of the successive stages may be kept substantially equal, and at a high multiplication factor. Other objects of my invention are to provide a multistage electron multiplier wherein such ions as may be formed are-directed to an impact area which is not concerned with electron multiplication and secondary emission; to provide such an electron multiplier without the necessity of using additional ion collecting electrodes, and where the ion collecting areas are of the same potential as the electron impacting areas; to provide an electron multiplier wherein the electrons originating in any area of impact are separated from the stream of impacting electrons so that there is'provided in the device a definite path followed by substantially all of the descendants of each impacting electron; and to provide an electron multiplier having stages therein or multiplying chambers wherein fields are produced tending to separate electrons from any ions that may be present, thus leading to an eificient, easily constructed device having a long useful life.

The invention possesses numerous other objects and features of advantage, some of which, together with the foregoing, will be set forth in the following description of specific apparatus embodying and utilizing my novel method. It is therefore to be understood that my method is applicable to other apparatus, and that I do not limit myself, in any way, to the apparatus of the present application, as I may adopt various other apparatus embodiments, utilizing the method, within the scope of the appended claims.

Referring to the drawings:

Figure 1 is a longitudinal sectional view of one preferred form of electron multiplier embodying my invention and practicing my method, together with a diagrammatic circuit showing one way in which the multiplier may be energized and operated.

modified form of the multiplier of this invention wherein the cup multiplying surfaces are curved.

Figure 5 is a schematic representation showing the positioning of the multiplying surfaces in the final stage in order to avoid their bombardment by ions.

In a prior Farnsworth application, Serial No. 82,888 filed June 1, 1936, now United States Patent No. 2,141,837 issued December 27, 1938, there is described a direct-current energized electron multiplier wherein each successive multiplying surface is positioned at substantially a right angle to the preceding and following surfaces, and where the electrons are removed from the active surfaces in a line substantially parallel to the plane of the surface. The present multiplier herein to be described, and illustrative of one embodiment of a structure utilizing the method herein disclosed, operates as far as positioning of surfaces is concerned in exactly the same manner as the multiplier previously described in Farnsworth Patent No. 2,141,887. However, surfaces are attached and form a continuation of the active surfaces in the present invention which change the conditions and fields operating on theelectrons and ions in their path between active surfaces in such a manner that ions produced are separated from the electrons.

It is believed that my invention, both as to means and method, may be more readily understood by direct reference to the drawings, beginning with Figure 1. An envelope i is provided at one end with a press 2, through which leads 4 pass supporting a thermionic cathode 5. This cathode may be directly or indirectly heated, as may be desired, as such indirectly heated cathodes are well known in the art. Immediately surrounding the cathode 5 is a space charge grid 5, and surrounding the space charge grid is a control grid I.

The thermionic structure thus described is merely one form of apparatus for providing an electron source which may be fully modulated with small control voltages. Any structure giving a high transconductance may, of course, be substituted. The high transconductance is a desirable although not absolutely necessary feature of multipliers of this type, the reason for which will be discussed later. The actual electron multiplying structure, next to be described, surrounds the thermionic structure.

The multiplier proper comprises a series of electrodes which may be considered as annular cups. The first of these cups is that bearing the reference character 9, having a flat bottom and a sidewall substantially perpendicular thereto. The thermionic structure projects through the central hole in the bottom of the cup, and is surrounded by a fine mesh screen l0, which is preferably a knitted fabric of extremely fine wire, such as that described in the copending application of Harry S. Bamford, Serial No. 108,553, filed October 31, 1936, now United States Pat-' ent No. 2,185,395, issued January 2, 1940. A tubular ring ll cooperates with the cup to support the screen It. The cup 9 has an inner surface which is treated to facilitate thesecondary emission of electrons, one form of cup which has been used in practice being made of pure silver, with the surface caesiated. The cup 9, screen It, and ring II are electrically continuous so that all operate atthe same potential. I

Supported above the cup 9 is the second multiplying electrode structure. This has a cylindrlcal portion I! which is preferably of the same diameter as the ring II and coaxial therewith, and an annular flange l4 parallel to the bottom surface of the cup 9. A screen l5, similar to the screen it, is supported above the opening of the cup 9 by means of a ring i8. As in the case of the preceding electrode structure, all of the portions of this second secondary emitting electrode areoperated at the same potential.

A third multiplying electrode structure surrounds the second, and comprises a cup I'l screen It, and supporting ring 20, and differs from that of the first stage principally in the larger size of the cup and the opening of the annulus.

Over the opening and the cup H a screen 24 is supported by a pair of rings 22 and 24, and over this screen the final secondary emitting electrode 25 is mounted, this last electrode being conveniently made in the form of a simple flanged disk. In forming the secondary emitting surfaces within the tube all of the electrode structures, including the screens Ill, l5, l9 and 2!, ordinarily become more or less secondarily emissive. This, in general, makes little difference, since the knitted screens are preferably formed of tungsten wire and their degree of secondary emission will be small in comparison to that of the caesiated silver surfaces. It may, however, be advisable to carbonize the screen 2|, as in certain connections this may be used as the final collector of electrons, and the other screens may also be carbonized if desired.

Figurejl shows one method of connecting the device. The incoming signal is applied through condenser 26 across an input impedance 2'! whose low potential end is connected to ground. Biases are applied to the space charge grid 6 and control grid I from a portion of the high potential source 29, in accordance with ordinary thermionic practice. Where maximum sensitiveness is desired, however, the positive bias on the space charge grid 6 may be made very low, or it may even be run at negative potential, so that it forms a virtual cathode whose emitted electrons all have very nearly the same velocity. cumstances an extremely small voltage change upon the control grid will serve completely to modulate the electron stream, and the fact that the modulated current is extremely small is unimportant because of the large current gains obtainable through multiplication.

The anode for this thermionic tube structure is the 01111.9, which is connected to a point positive with respect to the filament 5. Just how positive this point is depends upon the nature of the secondary emitting surfaces. In the caesiated surfaces of the type described the potential between the filament and cup may be anywhere from approximately 40 volts to approximately 300, depending upon the amount of multiplication desired and the function required of the tube. The lower potential given is that at which the secondary to primary electron ratio is unity,

while the higher potential mentioned is'that of maximum secondary to primary electron ratio. The second secondary emitting electrode i4 is connected to the source 29 to be positive again to the cup 9 while the cup I! is positive to the electrode M. The screen 2| is usually made very greatly positive to the cup l1, and the electrode 25 negative to the screen 2 i. The output resistor 30 is included in series with the electrode 2|, while the output power is shown as being withdrawn through the coupling condenser 3|.

The operation of the device is as follows: The modulated stream of electrons from the filament Under these cir-- attains its final velocity, due to the potential on the cup 9, by the time it passes through the screen l0. As soon as it enters the cup the stream is subjected to the field due to the greater positive potential on the screen l5, but because the electrons are already travelling at high velocity and because of the shielding effect of the walls of the chamber formed by screen I0, cup 9 and ring I I, the paths of the electrons are not greatly affected by this field, which results only in causing their paths to curve slightly as shown by the light arrow 32 of Figure 1. Substantially all of the electrons of the stream therefore hit the vertical wall of the cup 9, the arrows representing approximately their mean path.

The secondary electrons are emitted from the wall of the cup, but these electrons have, in general, a low velocity, and hence are accelerated by the field from the screen l5, most of them passing through the screen to hit the horizontal flange M. The same factors obtain in this case as did in the preceding one, the electrons gaining their full velocity by the time they enter the hollow electrode structure, and hence are only slightly deflected by the field from the screen I9. By similar reasoning the successive generations of electrons follow the light arrows of the figure, eventually impacting the electrode 25, from which the secondary electrons are drawn back to be collected by the screen 2! and its supporting structure, which operate at the most positive potential within the device.

In spite of the fact that every effort is made to operate the apparatus at as high a vacuum as possible, the vapor pressures of most of the good secondary emitters are appreciable and there will,

therefore, exist large numbers of gas or vapor molecules within the device. The high speed electrons used to promote secondary emission will ionize such molecules upon impact. In a single-stage device this is relatively unimportant, but the general type of device here shown can readily be made to give a multiplication factor of six per stage. The result is that even with a very small initial electron stream relatively large currents flow in the later stages of the device, the multiplication of the four-stage apparatus illustrated being approximately 1300 fold, so that an initial filament-emission current of 1 milliampere would represent an output of 1.3 amperes. It is clear that currents of these magnitudes will result in the generation of a large number of positive ions, and, as has already been stated, the efiect of bombardment by such ions upon secondary emissive surfaces is extremely destructive to such surfaces. The structure of the present apparatus is such that this bombardment does not affect those portions of the surfaces which are relied upon for secondary emission.

This effect is due to the fact that the fields within each of the multiplying chambers are such as to drive the resultant ions to a portion of the electron structure substantially at right angles to that from which secondary emission occurs. The heavy gas molecules within these chambers may be considered as at rest, since they have only the motion due to their thermal excitation. The momentum of the electrons, due to their small mass, is so low that the initial state of motion of the gas molecules is not perceptibly affected by the impacts which ionize them. The gas ions are therefore accelerated only by the field due to the positive charge upon the screen feeding to the next multiplying chamber, this charge repelling is, in fact, the case.

the positive ions and driving them against the bottom of the chamber, as illustrated by the heavy arrows 33 of Fig. 1. Destruction of the secondary emissive surface upon these portions of the electrode is actually an advantage, since it reduces the number of secondary electrons emitted by bombardment from the flow of positive ions, which latter secondary emission tends to destroy the practically perfect linearity of the device, and creates noise.

The multiplication in each stage is a function of the diiference oi. potential from the preceding emitter. For relatively small potential differences the ratio of secondary to primary electrons is practically a linear function of voltage. With increasing potential the secondary to primary electron ratio increases more gradually, to a fairly broad maximum, after which it falls off. It is seldom desirable to work any stage at less than unity gain, although this is theoretically possible and may at times be resorted to for special purposes. For the caesiated surfaces mentioned, unity gain occurs at about 40-volt, potential dif ference, while maximum gain is achieved at about 300 volts.

In order that the device have a linear response, one of the most important factors is that each stage operate at current saturation, and it is obvious that this is only possible where relatively high potentials are used, particularly in the later stages where the currents become appreciable. The establishment of space chargeswhich would prevent current saturation occurring is minimized by the fact that each chamber is larger than the preceding chamber. Since the apparatus operates at current saturation, its effective impedance is infinite, the current passed being a function only of the initial current and the over-all ratio of multiplication. To a first approximation, therefore, the power delivered by the device, for any given setting of the interstage voltages, is directly proportional to the magnitude of the impedance 30, and by increasing the voltage on the final stage, to compensate for the drop through the resistor 30, the power output of the device may be increased. The total voltage available in the final stage, under the most favorable conditions, is approximately double the direct-current potential of the circuit wherein the final or output impedance is located. Care should be taken, therefore, that at maximum output the screen 2| never swings far enough negative to permit a space charge to form beneath it.

All of the multiplier stages are ordinarily operated at ground potential with respect to the variable component of current, and if the source 29 is of high resistance the ordinary by-pass condensers are usually supplied to insure that this It is possible, however, to obtain additional effects by operating the tube with an impedance, typified by the impedance 35, in series with one of the multiplying electrodes.

One of the effects that may. thus be achieved is to make the device regenerative. impedance 35 is resistive, and the potential difference between the secondary emissive surface and the preceding stage is just greater than that required to give a one-to-one secondary to primary electron ratio, increased currents will cause a greater electron emission, which flowing through the impedancewill cause the potential of the electrode to rise. This, in turn, will increase the ratio of secondary electron emission,

' tending to cause a still further rise in secondary not of the two signals.

electrons. A very small impedance in this circuit is sufficient to produce this efiect. If, however,the voltage is just under that required for a one-to-one ratio, increase of infalling electrons will cause a potential drop in the reverse direction through the impedance 35, which decreases the-rate of electron emission and thus decreases the rate of gain in the stage, acting thus as a reverse feedback or degenerative circuit. With positive regeneration the stage into which the impedance is introduced has a squarelaw characteristic, and hence can be used for modulation and detection, as is well-known in the art. If the stage including the impedance be operated at a point where the ratio of secondary electrons to primary electrons varies alinearly with voltage, a less pronounced effect can be produced, and arrangement of this character may be used to compensate to some extent for space charge effects in tubes having large output.

Figure 2 shows one method of using the device as a modulator. Radio-frequency potential is applied to the grid 6 through a conventional input circuit 40. The modulating signal is applied to the cup 9, which is connected to the source 29 through the impedance 35, and the output current is therefore proportional to the prod- If the condenser 4|, shown bridged around the impedance 35, were omitted, both the modulated signal and the carrier could be applied to the grid, and modulation would occur owing to the square-law performance of this stage. The device could also be modulated, in accordance with known practice, by applying the carrier and the modulating frequencies to separate grids. In the modification shown in this figure, the output circuit 42 is shown connected in series with the electrode 25, and this electrode is made the most positive element within the tube and does not contribute to the multiplication. Otherwise the arrangement of Figure 2 is substantially the same as that of Figure 1.

Figure 3 shows an arrangement of electrodes which is applicable for use in image dissector tubes where both initial and final currents are very small and space charge effects do not, therefore, occur. The initial electrons from a photo sensitive cathode of a dissector tube enter through a scanning aperture 50 in a shield 5|, and impinge upon a conical secondary emitting electrode 52. From this point this course through the next three multiplying stages 54, 55 and 56, is the same as in the modifications already described. From the electrode 51, however, the flow is inward to the electrode 59, which is of smaller diameter than the preceding electrode, and from this point further electrodes may be added to any number of stages which seems advisable. Beyond the electrode 59 the arrangement of electrodes 55, 56 and 51 can be substantially repeated. In practice this has been carried as far .as ten multiplying stages, the enormous amount of multiplication involved being feasible because of the extreme minuteness of the input current, which may be of the order of amperes.

In strong contrast to the arrangement shown in Figure 3 is that of Figure 4, which is a schematic diagram of a structure adapted for highpower use. In this case the secondary emitting surfaces 60, GI and 62 are given curved sections. This renders them easier to manufacture, but does not distinguish sharply between those surtion.

faces which are subject to elettronic bombardment only and those which are subject to any ionic bombardment which may occur. The two effects will, however, take Place in general upon different portions of the surface owing to the curvatures of .the lines of force within the structures. The final collecting anode is the cup-like structure 65. This not only permits of a, large cooling surface and is therefore applicable to high power, but it also establishes what is very nearly a Faraday space within :the cup, so that the electrons are effectively collected, 1. e., they impart .their energy to the output circuit, even before they actually strike .the material of the cup. With the cup the most positive part of the circuit any secondary emission which may occur within it is re-collected by the cup and does not cause trouble.

Figure 5 shows the cross-sectional outline of an electrode 61 which is designed still further to decrease the probability of an ion hitting the emitting surface. The outer wall of the cup slants inward, so that ions generated in the space between the final screen 69 and the final anode 10 will not hit the emitting surface even though they may have a material velocity component laterally outward. As in Figure 1 .the electron paths are shown by light arrows and the ion paths by a heavy arrow.

There is another effect which can be obtained with electrodes having the cross-section diagrammed at 61 in Figure 5, which may be useful when such electrodes are used as modulators. Under these conditions the paths of low speed primary electrons are much more curved than when the electrode is at relatively high potential; emission is more strongly localized at the lip of the cup; and space charge effects on the incoming primary electrons may occur, cutting down the secondary emission ratio more than linearly with voltage. With the electrode shaped as,is electrode 61, high speed electrons hitting nearer the angle of the cup will generate secondary electrons in a somewhat shielded position, so that a compensating space charge condition will ensue, resulting in substantially linear opera- It will be clear that there are here illustrated only relatively few of many possible modifications of my invention, which is applicable to almost any service which calls for the so-called static or direct-current type of multiplier. In practice it has proved to give the most consistent results of any of the types of multiplier of this character with which I am familiar, both for extremely small currents, as in the arrangement of Figure 3, and for relatively high powers, and while I have-not as yet built amplifiers embodying this principle in sizes of over a few hundred watts output .the extreme flexibility and straightforwardness of design indicate that it can be made to deliver extremely large powers. One of its primary advantages is that there is no theoretical limit .to the amount of power which can pedance of 1,000 ohms, represents about thirty I kilowatts controlled by the initial grid swing of two volts for two tubes working in push-pull. In

such an utilization all of the secondary emitting electrodes may be kept at ground potential, and since there are no intermediate tuned circuits to cause frequency discrimination, the advantages of such an arrangement are obviously very great, and these advantages may beutilized fully, with a tube of long life, because the arrangement of opposite said side wall being open, means for directing electrons through said screen to impact said surface, and means for removing secondary electrons through the opening in said chamber.

2. An electron multiplier having an envelope containing an electron multiplying chamber comprising a surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, an electron-permeable screen parallel to said surface, a conductive side wall on one side of said surface, means for directing electrons through said screen .to impact said surface, means for removing secondary electrons at one side of said surface, and means for collecting ions at the opposite side of said surface.

3. An electron multiplier having an envelope containing an electron multiplying chamber comprising a surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, an electronpermeable screen parallel to said surface, a conductive side wall on one side of said surface, means for removing secondary electrons from said surface in a path generally parallel to said surface in one direction, and means for removing ions in said chamber along a path differing in direction from said electron path.

4. An electron multiplier comprising an envelope containing a surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, means for directing primary electrons against said surface along a path substantially perpendicular thereto to create secondary electrons by impact, means for removing said secondary electrons from said surface along a path substantially parallel thereto, and means for preventing ions from impacting said surface.

5. An electron multiplier comprising an envelope containing a surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, means for directing primary electrons against said surface along a path substantially perpendicular thereto to create secondary electrons by impact, means for removing said secondary electrons from said surface along a path substantially parallel thereto, and means for removing ions from said primary electron path before they reach said surface.

6. An electron multiplier comprising an envelope containing a surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, means for directing primary electrons against said surface along a path substantially perpendicular thereto to create secondaries by impact, means for removing said secondaries from said surface along a path substantially parallel thereto, and fleld-,

determining members positioned to produce a field operating on ions in said primary path to reduce the probability of their reachirg said surface.

7. An electron multiplier comprising an envelope containing a surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, a collecting electrode for said secondary electrons positioned at one side of said surface and in a plane at an angle to said surface, means for directing electrons emitted from said surface on to said collector, and means forming a continuation of saidsurface extending substantially perpendicularly to the plane of said collector for intercepting positive ions.

8. An electron multiplier comprising an envelope containing an electron collecting electrode, an unheated electron emitting electrode spaced from said collecting electrode and having an active surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, said active surface having an angle to the plane of said collecting electrode of not less than 90", whereby ions moving toward said surface tend to miss it, and an electron-permeable accelerating electrode intermediate said electrodes.

9. An electron multiplier comprising an envelope containirm an electron multiplying chamher having a substantially rectangular cross section, one wall of said chamber being open, the opposite wall being solid, one of the remaining walls being electron-permeable and the wall opposite said electron-permeable wall being capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, and means for introducing electrons into said chamber through said electron-permeable wall.

10. An electron multiplier having a plurality of chambers according to claim 9, and where the open 'wall of one chamber is presented to the electron-permeable 'wall of the next chamher, and means for introducing primary electrons into thefirst chamber.

11. An electron multiplier comprising an envelope containing a plurality of annular cups, the cross section of said cups being substantially rectangular with one wall open, the opposite wall closed, one of the remaining walls being capable positioned cups through the electron-permeable wall thereof, and means for collecting electrons emerging from the open wall of the last chamher of said series, said chambers being energized to increasingly higher potentials from the first to the last chamber.

12. An electron multiplier comprising an envelope containing a plurality of annular cups, the cross section of said cups being substantially rectangular with one wall open, the opposite wall closed, one of the remaining walls being capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, and the last wall being electron-permeable, said cups being serially positioned with the open wall of one cup presented to the electron-permeable wall of the next cup, means for introducing electrons into the first chamber of the serials through the electron-permeable wall thereof, means for collecting electrons emerging from the open wall of the last chamber of said series, said chambers being energized to increasingly higher potentials from the first to the last chamber, and means for varying the introduced electrons.

13. An electron multiplier comprising an envelope containing a plurality of annular cups, the cross section of said cups being substantially rectangular with one wall open, the opposite wall closed, one of the remaining walls being capable of emitting secondary electrons at a-ratio greater than unity upon electron impact therewith, and the last wall being electron-permeable, said cups being serially positioned with the open wall of one cup presented to the electron-permeable wall of the next cup, means for introducing electrons into the first chamber of the serially positioned cups through the electron-permeable wall thereof, means for withdrawing electrons from the last chamber through the open wall thereof, and means for collecting the withdrawn electrons, said chambers being energized to increasingly higher potentials from the first to the last chamber.

14. An electron multiplier comprising an envelope containing a cylindrical surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, means within and substantially axial of said cylinder for producing a stream of electrons, means for directing electrons radially to impact said surface, and means for withdrawing the secondary electrons produced by said impact along paths substantially parallel to the axis of said cylinder.

15. An electron multiplier comprising an envelope containing a cylindrical surface capable of emitting secondary electrons at a ratio greater than unity upon electron impact therewith, means for producing an electron stream travelling along the axis of said cylinder, and a coneshaped electrode positioned coaxially with said cylinder and positioned with its apex presented to said stream whereby secondary electrons are produced by impact therewith, and means for directing said secondary electrons along radial paths to impact said surface.

16. In an electron multiplier, a succession of electron permeable screens, each successive screen beingpositioned with its elements substantially perpendicular to the elements of the preceding screen, an electrode having a surface capable of emitting secondary electrons at a ratio greater than unity upon the impact of primary electrons therewith positioned intermediate each pair of successive screens and forming a chamber therewith, connections adapted for the energization of said screens at different potentials, and means for projecting primary electrons through one of said screens.

17. In an electron multiplier, a succession of electron permeable screens, each successive screen being positioned with its elements substantlally perpendicular to the elements or the preceding screen, an electrode having a surfaceperpendicular to the elements of the preceding screen, an electrode positioned intermediate each pair of successive screens and having surfaces facing each screen, the surface facing the earlier screen of the series being capable of emitting secondary electrons at a ratio to impacting primary electrons which. is greater than unity, con nections adapted for the energization of said screens at different potentials, and means for projecting primary electrons through one of said screens.

RICHARD L. SNYDER.

Images