US2883531A

US2883531A - System using counter tube coder

Info

Publication number: US2883531A
Application number: US518312A
Authority: US
Inventors: Carl F Anderson
Original assignee: PACKARD BELL Co
Current assignee: PACKARD BELL Co; PACKARD-BELL Co
Priority date: 1955-06-27
Filing date: 1955-06-27
Publication date: 1959-04-21
Anticipated expiration: 1976-04-21

Description

5 Sheets-Sheei'. 1

Imam/70x2 mxvbim o zEbmJmw mbmm Q5 c. F, ANDERSON SYSTEM usmc COUNTER TUBE CODER April 21,v 1959 Filed June 27, 1955 United States Patent SYSTEM USING COUNTER TUBE CODER Carl F. Anderson, Los Angeles, Calif., assignor to Packard-Bell Company, Los Angeles, Calif., a corporation of California Application June 27, 1955, Serial No. 518,312

9 Claims. (Cl. 250-27) The present invention relates to improved means and techniques for developing a series of time spaced pulses of the character, for example, used in so-called IFF systems wherein it is desired to produce different combinations of accurately spaced pulses for identification purposes.

Systems of this character have heretofore been proposed and they invariably include a so-called delay line as an essential part in achieving the spacing between pulses. One such prior art system involves the use of a pulse generator driving a delay line in such a manner that the total delay time which is achieved by the delay line is the total time required between the training pulses of the generated coded train. The number of taps required on the delay line is determined by the number of information pulses that are desired in the pulse train. The input, output and various taps on the delay line are fed to a code connecting network that has a common output, and used to display the code train. A prior art system of this type has several disadvantages in that (1) all of the pulses collected from the delay line are of diiferent amplitudes and shape; (2) the amplitude of the pulses in the train change as the pulses are switched in or out of the pulse train, i.e. code structure; and (3) it is a difficult and expensive job to build a multitapped delay line to the time tolerances required by such a coder.

Another prior art system involves the use of a sine wave oscillator synchronized with the line drive pulse generator with the oscillator serving to produce the actual code pulses, with an accompanying delay line used to supply only timing information. However, certain problems are involved in such system requiring relatively complex circuitry with its attendant probability of failure.

In accordance with the present invention, such prior art systems are greatly simplified and made more reliable by the use of a so-called beam switching tube which eliminates the necessity of a delay line.

It is therefore a general object of the present invention to provide improved means and techniques for developing a train of pulses.

A specific object of the present invention is to provide a system of this character which avoids the use of delay lines.

Another specific object of the present invention is to provide a system of this character which is relatively simple and reliable and which is easily adjusted.

Another specific object of the present invention is to provide a system of this character allowing convenient switching in and out of pulses in a coded train of pulses without affecting the amplitude or shape of the remaining pulses.

The features of the present invention which are believed to be novel are set forth with particularity in the appended claims. This invention itself, both as to its organization and manner of operation, together with further objects and advantages thereof, may be best understood by reference to the following description taken in connection with the accompanying drawings in which:

Figure l is a schematic representation of a system embodying features of the present invention;

Figure 2 is a transverse sectional view through a beam switching tube that is connected in the circuitry illustrated in Figure 1;

Figure 3 represents additional circuitry which may be added to the circuitry illustrated in Figure 1 for obtaining information pulses in 1.45 microsecond steps;

Figure 4 represents the schematic form of another modification of the present invention; and

Figure 5 is a block diagram of the apparatus illustrated in Figure 1.

Briefly, the apparatus shown in Figures 1 and 5 functions as follows.

The positive trigger 10 applied to the input terminal A is applied to the input grid of the bistable multivibrator code gate generator 12, causing it to change states. The resulting negative gate 14 appearing on the connected cathodes of the tubes VIA and V1B is applied to the control grid of the oscillator clamp tube V2A, causing a cessation of the space current through tube V2A. This sudden cessation of current through tube V2A and the inductance L1 in the oscillator 16 causes the oscillator 16 to be shocked into oscillations which are sustained by the Hartley type oscillator 16. The resulting sine waves have two phases which are degrees apart and are applied to the control grids of tubes V3A and V3B which comprise a push-pull driver stage 20. The outputs from the driver stage 20 are applied to alternate connected grids, i.e. the odd and even grids, respectively, of a beam counter tube 22, which is illustrated in Figure 2.

As the potential of the grids of the counter tube 22 are progressively lowered, the beam in the counter tube is caused to shift conduction from target to target in rotation at a frequency that is two times the frequency of oscillations developed in the oscillator stage 16.

Since the voltage developed on each tube target 22T is substantially independent of each other, a

switch

24, 25, 26, 27, 28 and 29 in series with a

corresponding isolating diode

31, 32, 33, 34, 35 and 36, respectively, and placed between the differentiated target output voltage, and a common collecting resistor R17 allows the coding or selection of pulses to appear on a collecting resistor R17, so as to develop a series of positive pulses, as represented by the pulse train 40 and appearing on the output terminal 42.

The output of the beam tube target 22T1 is dilfereni tiated and then applied through the shut-01f pulse amplifier stage 46 to the connected cathodes of the gate generator 12 to trigger the bistable gate generator. This will cause .the gate generator to assume its original state, and the negative gate on the grid of oscillator clamp tube is thereby removed, allowing such clamp tube to conduct and to damp out the oscillations in the tank circuit including inductance L1. The counter thereby comes to rest with the beam on the last target and a circuit is then ready to repeat its operation upon the application of another positive trigger 10.

The tube 20 may be of the type supplied by Burroughs Corporation and known as the type MO-10. The MO-lO is a magnetron type vacuum tube in which an electron beam is formed by crossed electric and magnetic fields. A small permanent magnet (not shown) encloses the tube envelope 22E (Figure 2) and supplies the latter field. The electron beam designated by the line 22B is formed between the central cathode 22C and one of ten circumferential positions, each of which contains the three electrodes: (a) spade 2281-22810, inclusive, to form and lock in the beam, (b) targets 22T1-22T10, inclusive, to render a useful output, and (c) grids 22G1-22G10, inclusive, to switch the beam to the next posit-ion.

The counter tube MO-lO is effectively a ten position electronic switch with ten separate outputs with-in one small envelope. The MO-lO can be preset to any one of its ten positions and the beam can remain indefinitely on one position or it can be advanced continuously at rates which exceed one megacycle.

The MO-10 as used commercially has an eleven stable state, namely a so-called clear condition in which the beam does not exist. One method by which the tube can be placed in this clear condition is the momentary interruption of the spade and target positive supply voltages. From the clear condition the electron beam can be formed on a particular position by momentarily reducing the voltage of the spade at that position to a value approximately equal to the cathode potential. The circuitry described in Figure 1 is now described in detail as follows. The input terminal A is connected to the control grid of tube V1A through coupling condenser 50. The grid of tube V1A is returned through resistance R2 to a source of 40 volts. The cathodes of tubes VIA and V1B are each connected through the output resistance R3 to the same 40 volt source. The anodes of tubes VIA and V1B are connected respectively through resistances R4 and R5 to the +125 volt source. Condenser Cl connects the anode of tube V1A to the grid of tube V1B. Resistance R1 is connected between the grid and cathode of tube V1B.

The clamp tube V2A has its anode connected directly to the +125 volt source, its grid connected to the cathodes of tubes V1A, V1B and the cathode of tube V2A is returned to ground through the tapped inductance L1, which is tuned by the adjustable condenser C2, connected in shunt therewith.

The oscillator tube V2B has its cathode connected to a 125 volt source through a pair of serially connected

resistances

52 and 54, and a junction of these

resistances

52, 54 is connected through resistance 56 to the control grid :of tube V213. The anode of tube V2B is connected to the +125 volt source through resistance 58. Con-- denser 60 .couples the control grid of tube V2B to the ungrounded terminal of the tank circuit of the oscillator which includes the elements L1 and condenser C2.

Two output are derived from the oscillator stage 16' and applied in proper phase to the push-pull drive-r stage 20. One of such two outputs is derived from the anode of tube V2B and is coupled by condenser 62 to the control grid of tube V3A. A second one of such two outputs is derived from the oscillator cathode circuit and is applied to the control grid of tube V313 through a net work which includes the condenser 66 and adjustable resistance R6. Condenser 66 is connected between the junction point of

resistances

52 and 54, :on one hand, and the control grid of tube V3B on the other hand. A-djust-' able resistance R6 has one of its terminals connected to the tap on coil L1 and its other terminal connected to the grid of tube V3B. R6 serves to control the amplitude of the signal applied to the grid of tube V3B.

Tubes V3A and V3B have their cathodes intercon uected and returned to ground through a common resist ance 70 which is shunted by the condenser 72. The control grids of tube V3A and V3B respectively are connected to a 125 volt source through corresponding re sistances 74 and 76. The cathodes of tubes V3A and V3B are connected to such 125 volt source through the common resistance 78. The anodes of tubes V3A and V3B are each connected respectively to the corresponding outside terminals of the resistance R7 through corresponding resistances 80 and 82. An adjustable tap on resist-- ance R7 is connected to the +125 volt source. The junction point of resistances 80 and R7 is connected to ground through the shunt connected resistance 84 and condenser 86. Likewise, the junction point between resistance R7 and resistance 82 is returned to ground through the shunt connected resistance 88 and condenser 90. The anode of tube V3A is directly connected to each of the odd numbered grids of the counter tube 22, while, likewise, the

anode of tube VSB is connected to each one of the even numbered grids of tube 22.

Each one of the spades of tube 22, namely the spades 2281-22810, is connected to a positive 125 volt source through corresponding resistances 91-100. It is noted that the spade 2281 is connectible to ground through switch S1, and that the spades 2259 and 22310 are interconnected by resistance 99A.

One terminal of each of the target electrodes 22T1- 22TH) is connected to the same 125 volt source through corresponding resistances 101110. The other terminals of targets 22T9 and 22T10 remain unconnected as shown. The other terminals of the targets 22T1--22T8 are con-- nected respectively to one terminal of the diode rectifiers 30-37 through corresponding condensers C5C12, the other terminals of the

rectifiers

30 and 37 being connected directly to the ungrounded terminal of the output resistance R17, while the other terminals of the rectifiers 31- 36 are connectible through corresponding switches 24-29 to such ungrounded terminal of the output resistance R17.

The pulses developed on the targets 22T1--22T8 occur at diflferent times as indicated by the laterally spaced negative pulses 111118, appearing on the corresponding targets.

The negative pulse 111 is applied through condenser C3 and diode to the control grid of the amplifier tube 46. The junction point of the diode 120 and condenser C3 is returned to ground through resistance R8 to form of differentiating network. The control grid of tube 46 is returned to ground through resistance 122 and the oathode of tube 46 is returned to ground through resistance 124 which is shunted by condenser 126. The anode of tube 46 is connected to a positive 125 volt source through the resistance 128. The anode of tube 46 is connected through condenser and diode 132 to the cathode of tubes V1A, V1B. The junction point of the

elements

130 and 132 is returned to a '-4() volt source through resistance 134.

The operation of the circuit as shown in Figure l is as follows.

Before a positive trigger is applied to the point or input terminal A of the gate generator, the gate generator and the oscillator clamp are initially subjected to the following conditions, namely, tube VlA is at cutofii and V1B is drawing maximum current. This is the normal steady state condition, since tube V1 which comprises the sections VIA and V1B is a monostable multivibrator and the grid resistor R1 of V1B is returned directly to its cathode while the grid resistor R2 of V1A is returned to -40 volts. The value of R3, in the cathode of tube V1, is such that the DC. voltage at point B, i.e. at the cathodes of tubes VIA and V1B, is zero with respect to ground. Since the grid of the oscillator clamp tube V2A is connected directly to this point, it is at the same D.-C. potential and tube V2A draws approximately 15 milliamperes of current. Under these conditions, the cathode of tube V 2A presents a low impedance across the tank circuit of the Hartley oscillator V2B, thus preventing it from oscillating.

When a positive trigger voltage is applied to point A, tube VIA is turned on and tube V1B is shut off. The multivibrator remains in this condition, unless the circuit is influenced by an external voltage, for a period of time which is determined by the values of condenser C1 and resistance R1. The time constant C1R1 is set to be about 30 microseconds, which is approximately the time required for one and one-half code trains.

Tube V1A draws less current than tube V1B because the plate resistor R4 is larger in value than R5. Thus, the Voltage at point B becomes negative with respect to ground. A negative pulse is produced in this manner at the grid of tube V2A of sufiicient amplitude to shut off tube V2A. The current which is drawn by tube V2A through the oscillator coil L1 is suddenly reduced to zero. The collapsing field in coil Ll sets up a shocked oscillatory condition in the tank circuit L1, C2. The oscillation always starts in the negative direction since the energy in the tank is reducing. Resistance R6 in the cathode circuit of tube VZB is adjusted to the point where tube V2B maintains the oscillations at a constant amplitude. In other words, tube V2B supplies just enough energy to the tank circuit to overcome circuit losses.

Tube V2B is also used as a phase inverter. The cathode and plate outputs are adjusted for equal amplitudes by selecting the proper load values for the two circuits.

Without a signal applied to the grids of the push-pull driver tube V3A, V3B, the resistance network in the plate circuit of tube V3 is designed so that approximately +30 volts with respect to ground is applied to the anodes of tube V3. This is done so that the grids of the beam switching tube 22, which are directly connected to the anodes of tube V3, are at the proper bias point. To make it possible to pull the plates of tube V3 below ground and also to establish the proper operating point for tube V3, the cathodes of tube V3 are returned to a negative voltage. The grid bias on tube V3 is set so that under no signal conditions the tube is at or slightly beyond cutoff. Since only the positive portion of the phase inverted sine waves produce a change in the plate circuit of tube V3, the anode of tube V3A is the first pulled below ground,- followed one-half cycle by the anode of tube V3B. If the anodes of tube V3 are observed simultaneously, their combined outputs appear as a full wave, rectified, sine wave. The anode of tube V3 is connected directly to all the odd numbered grids and the anode of tube V313 is connected directly to all of the even numbered grids of the beam switching tube, and because of the full-wave rectification mentioned above, the beam switching tube steps to its progressive positions at twice the rate of the fundamental frequency of the oscillator.

When the coder is initially turned on, it is necessary to form the beam in the beam switching tube 22 on the proper spade. This is accomplished by closing switch $1, a momentary contact switch, which is located at the top of the spade load resistor network. This reduces the voltage on number one spade to zero and forms the beam in the number one position.

.When the large negative-going signal from the anode of tube V3A is applied to the odd numbered grids of the beam switching tube 22, the stable condition existing at position one is upset and the beam moves to position two. Immediately thereafter, the anode of tube V3B applies a large negative-going signal to the even grids of the beam switching tube and the beam moves to position three. In this manner, the beam is progressively stepped around the tube. Since the beam switching tube which is used has ten positions and since the coder, as described, requires an eight-position'tube, it is necessary to make the ten-position tube operate as if it were an eight-target tube. This is accomplished by vconnecting together spades nine and ten through the proper resistance value. As the beam strikes each spade, the voltage on the spade is decreased due to the current drawn by the spade through its load resistance. As the beam strikes spade number nine, it does not find a stable condition since the voltage on, spade number ten is also decreasing because of the resistance that ties the two together. The beam, therefore, moves to position number ten. It does not stay in this position since all of the even numbered grids are still pulled below ground by the signal from the anode of tube V3B. In this manner, the beam moves on to position number one. It is at this point that the progress of the beam around the tube must be stopped since it has passed through all the required positions for generating a com.- plete pulse train. As the beam strikes each target, the voltage on the target is reduced for the same reason mentioned in connection with the spade voltage. The leading edge of the negative pulse produced as the beam strikes target number one is differentiated by condenser C3 and R8, and applied through the coupling diode to tube V5. Tube V5 phase inverts and amplifies this pulse and applies it through the coupling diode to point B. The effect of this positive pulse applied to point B is l) to force the monostable multivibrator back to its normal state which, of course, removes the negative gate from point B, and (2) to drive the grid of tube V2A into the positive region to cause the oscillator tube V2B to stop oscillating immediately. The coder now has completed one complete cycle and is ready for the next trigger pulse.

The output or code pulses are generated and selected in the following manner.

The framing and information pulses are produced as the beam leaves each target and not when the beam first strikes the target. Since the framing pulses and each code pulse are generated in exactly the same manner, the generation of the first framing pulse only is described. As the beam leaves target number one, the rising voltage on number one target is difierentiated by condenser C5 and resistance R9. The resulting positive pulse is passed by the collecting diode 30 and appears across the common load resistance R17. The information pulses are inserted or removed from the code structure by operation of the proper code selecting switches 24-29.

The spacing between pulses in the output train 40 is controlled by the Hartley oscillator tuning condenser C2 and resistance R7, the adjustable voltage divider in the anode circuit of tube V3. By observing the first (framing pulse and the second information pulse (on an externally synchronized oscilloscope), condenser C2 is set to obtain 5.8 microseconds between the leading edges of these two pulses. The first framing pulse and the first information pulse are then displayed on the oscilloscope, and resistance R7 is adjusted to obtain 2.9 microseconds between the leading edges of this pair of pulses. Condenser C2 sets the frequency of the Hartley oscillator, which controls the spacing between even pulses and the spacing between odd pulses. Resistance R7 sets the switching level between the odd and even grids of the beam switching tube 22, and this controls the spacing between the odd and even pulses. Resistance R7 compensates for any pulse shape or amplitude variation between the outputs of tubes V3A and V3B.

Thus, using the circuit shown in Figure 1, any one of six information pulses corresponding to switches 24--29 may be inserted in or removed from the pulse train which is bracketed by the first and last pulses, namely the framing pulses developed respectively from target one and target eight. These pulses, as indicated above, are separated by time interval of 2.9 microseconds. In order to decrease the time spacing between selectable pulses to 1.45 microseconds, the circuitry shown in Figure 3 is added to the circuitry shown in Figure 1 by making additional connections to the targets of tube 22, as indicated.

The pulses which are developed in Figure 3 are applied to the same collecting or output resistance R17 which is identical to the resistance R17 shown in Figure 1. In this case, all of the pulses developed in Figure 3 are delayed 1.45 microseconds by the delay line 202 so that these pulses fall between, in time, the pulses developed by the apparatus shown in Figure l.

The circuitry shown in Figure 3 includes seven selecting switches 223-429, which serve to selectively connect the control grid of tube V5 to a corresponding one of the targets one-seven of the beam tube 22 through corresponding diodes 230-236 and through corresponding condensers C25C31, which condensers are associated respectively with a corresponding resistance R29R35 to form a corresponding difierentiatin-g network.

The tube V5 has its anode connected to a volt source and has its control grid returned to ground through resistance 240. 'Ilhe cathode of tube V5 is connected to ground through the coil 202L of the delay line 202, and the resistance 242. The delay line 202 includes the shunt capacitor 2020 which may be the self-capacity of the coil 202L. The junction point of the delay line 202 and resistance 242 is coupled to the ungrounded terminal of resistance R17 through coupling condenser 244, which is connected in shunt with resistance 246. Thus, when the circuitry shown in Figure 3 is added to the coder, information pulses are available in 1.45 microsecond steps. The first framing pulse is fed to the grid of tube V through the first selecting switch which is closed. Such pulse is delayed 1.45 microseconds by the delay line 292 in the cathode circuit of tube V5, and reinserted into the common code collecting network where it becomes the first information pulse in the code structure. Eaclh succeeding 2.9 microsecond pulse is delayed and reinserted into the code structure in the same manner. The last training pulse developed on target eight is not fed through the delay circuit, since if it were reinserted, it would fall outside of the last framing pulse produced, using the apparatus shown in Figure 1. There are now thirteen selectable information pulses available, instead of the original six. Of course, all of the thirteen pulses do not have to be used, and any that are not required may be switched out of the circuit.

The circuitry shown in Figure 4 is able to accommodate two trigger sources, both feeding information into a common beam switching or counter tube 22, having the aforementioned targets one-eight. The gate generators 12A and 12B are each exactly as the gate generator 12 described in Figure 1, and the points or terminals A, B and C correspond to the terminals A, B and C in Figure 4. The terminals B and B are each connected to the control grid of the clamp tube V2A but, in this case, isolating diodes CR1 and CR2 are provided so that one gate generator does not trigger the other gate generator. Another slight modification is that resisance R18 is connected between the control grid and the positive source of voltage so as to again normally maintain the clamp tube V2A conducting in its quiescent state. In this case the value of the resistance R18 is such that, under no-signal conditions, the grid current drawn by the tube V2A through this resistance holds the grid of tube V2A at zero volts with respect to ground. The points A and B in Figure 4 have the same significance that they have in Figure 1, and it is noted that the point C is also designated in Figure 1 and is the output from the anode of tube VlB. The voltage on the points C and C is applied in Figure 4 through corresponding coupling condensers 301 and 302 to corresponding biased diodes CR3 and CR4, which diodes are normally biased negatively so as to disable the code selecting network. In order to render such code selecting networks, either N or N in Figure 4, it is necessary that positive voltage of sufficient intensity be applied to the diodes CR3 and CR4, which are maintained at a negative potential through

resistances

303 and 304. Such enabling positive voltage is derived from the gate generators 12A or 12B, as the case may be. In other words, all of the input load resistances for the collecting diodes in Figure 4 are removed from ground and returned to -40 volts. This voltage disables the collecting network until the 40 volts is removed. When either one of the gate generators 12a or 12b, as the case may be, receives a positive trigger, the positive pulse that is generated at point C or C is applied to the proper disabling circuit. The positive pulse applied allows the diode CR3 or CR4, as the case may be, to clamp the disabling circuit to ground, and thus, the code collecting network in question is ready for operation. The coder is now able to reply with either of the two preset codes, depending on which gate generator is triggered. The output of each code selecting network N and N is applied to a common output resistance R17.

While the particular embodiments of the present invention have been shown and described, it will be ob vious to those skilled in the art that changes and modifications may be made without departing from this invention in its broader aspects and, therefore, the aim in the appended claims is to cover all such changes and modifications as fall within the true spirit and scope of this invention.

I claim.

1. In a pulse transmission system of the character described for transmitting a series of pulses separated in integral fractions of time corresponding to the periodicity of oscillations, an electronic switching tube having a plurality of electrodes which are progressively swept by an electronic beam, a high frequency oscillator developing said oscillation and coupled to said tube for moving said 'beam in synchronism with the oscillations developed in said oscillator, a common output circuit, selectable switching means connecting a corresponding one of said electrodes to said common output circuit, second selectable switching means coupled to a corresponding one of said electrodes, and delay means interposed between said common output circuit and said electrodes.

2. In a pulse transmission system of the character described for transmitting a series of pulses separated in integral fractions of time corresponding to the periodicity of oscillations, an oscillator producing said oscillations, sequentially operated electronic switching means operated in timed relationship with said oscillations, means for operating said switching means through a cycle, said switching means having a plurality of output circuits, a common output circuit, first selectable switching means connecting a corresponding one of said output circuits to said common output circuit, a second set of selectable switching means connecting said output circuits to said common output circuits, a first pulse responsive means for controlling said operating means, a second pulse responsive means for controlling said operating means, said first pulse responsive means, when energized, being effective to render only one of said selectable switching means effective to couple said output circuits to said com mon output circuit, and said second pulse responsive means, when energized, being effective to render the other one of said selectable switching means elfective to couple said output circuits to said common output circuit.

3. In a pulse transmission system of the character described for transmitting a series of pulses separated in integral fractions of time corresponding to the periodicity of oscillations, electronic switching means comprising a tube having a rotatable cathode beam engageable in succession with a plurality of electrodes, an oscillator producing said oscillations and coupled to said tube for producing movement of said cathode beam in timed relationship with respect to the oscillations developed in said oscillator, pulse responsive means for initiating operation of said oscillator, said pulse responsive means including means for assuring a predetermined phase of the first half one of the oscillations developed by said oscillator, and means coupled to one of said electrodes for interrupting the oscillations developed in said oscillator.

4. A system as described in claim 3 including a common output circuit, and selectable switching means between a corresponding one of said electrodes and said output circuit.

5. A system as described in claim 4 including a second set of selectable switching means, delay means, and said second set of selectable switching means and said delay means being connected between said output electrodes and said common output circuit.

6. In a pulse transmission system of the character described for transmitting a series of pulses separated in integral fractions of time corresponding to the periodicity of oscillations, a gate generator having an input circuit responsive to a pulse for initiating a gating voltage, an oscillator producing said oscillations, a clamping circuit for said oscillator and coupled thereto for normally rendering said oscillator ineffective to develop oscillations, said clamping circuit being coupled to said gate generator and responsive to said gating voltage, said clamping circuit being energized by said gating voltage to allow said oscillator to oscillate, an electronic switching tube having a cathode beam movable in succession into engagement with a plurality of electrodes, means coupling said oscillator to said tube to move said cathode beam in synchronism with the oscillations developed in said oscillator, and means coupled between one of said electrodes and said gate generator for terminating said gating voltage to thereby restore said oscillator to its normally non-oscillating condition.

7. A system as described in claim 6 including a common output circuit and selectable switching means connected between a corresponding one of said electrodes and said common output circuit.

8. A system as described in claim 7 including second selectable switching means, delay means, said second selectable switching means and said delay means being interposed between said electrodes and said common output circuit.

9. In a pulse transmission system of the character described for transmitting a series of pulses separated in integral fractions of time corresponding to the periodicity of oscillations, an electronic switching tube having a cathode beam movable in succession into engagement with a series of electrodes, an oscillation network for developing said oscillations and coupled to said tube for moving said cathode beam in timed relationship with the oscillations developed in said oscillator, means normally rendering said oscillation network ineflective to develop oscillations, at common output circuit, a first set of selectable switching means interposed between said electrodes and said common output circuit, means normally rendering said first set of selectable switching means ineffective to connect said electrodes to said common output circuit, a second set of selectable switching means connected between said electrodes and said common output circuit, means normally rendering said second set of selectable switching means ineffective to connect said electrodes to said common output circuit, a first pulse responsive means for controlling said oscillation network, a second pulse responsive means for controlling said oscillation network, means coupled between said first pulse responsive means and the first-mentioned rendering means for disabling the first-mentioned rendering means to thereby render said first selectable switching means effective to connect said electrodes to said common output circuit, and means coupled between said second pulse responsive means and the second mentioned rendering means for rendering said second mentioned rendering means inefiective to thereby render said second set of selectable switching means efiective to connect said electrodes to said common output circuit.

References Cited in the file of this patent UNITED STATES PATENTS 2,097,392 Finch Oct. 26, 1937 2,404,920 Overbeck July 30, 1946 2,418,521 Morton et a1. Apr. 8, 1947 2,536,035 Cleeton Ian. 2, 1951 2,539,623 Heising Jan. 30, 1951 2,690,507 Woods-Hill et a1. Sept. 8, 1954 2,711,532 Slusser June 21, 1955 2,848,647 Kuchinsky et al. Aug. 19, 1958