NO118088B - - Google Patents

Description

Electronic code connection system.

This invention relates to an electronic code communication system especially for railways, of the kind described in American patent no. 2,629,088.

The electronic code connection system according to the invention is generally of the type intended for the transmission of electricity

ke control signals and indication signals between a manager's office and at least one field station, where a transmitter device is placed at the manager's office 'equipped to send successive cycles of control signals that have a mutual time gap, to the field station or each field station in turn during a cycle, while at the field station or at each field station a transmitter device is placed designed to send indication signals back to the manager's office in response to control signals being received at the field station which have been sent to the relevant

de field station from the management office. A main characteristic of the invention is that the field station or each field station has a signal storage device in the form of an electronic shift register, a circuit for setting the steps in this register for storing a sequence of indication signals, as well as a further circuit controlled by control signals received from management office to tap out the stored signals from the register, so that this taping takes place timed in relation to the control signals received at the field station.

The advantages of the system according to the invention are, firstly, that when using an electronic storage device, high speed is ensured both when sending and receiving, despite the fact that the controlled devices to which the sent information relates are rather slow to respond; secondly, it is achieved by using a shift register, so that information signals can be received and sent at the field station one at a time; and thirdly, by sending these signals one at a time, it is achieved that one only needs to use a single storage device, both for incoming and outgoing information, because each indication signal that is sent is replaced in the register by a received control signal.

Other purposes and purposes and characteristic features of this invention will be apparent from the following description of an exemplary embodiment, referring to the attached drawings, where: Fig. 1 is a simplified block diagram showing the general structure of the system according to the present invention. Fig. 2 is a simplified time chart showing the general code structure for the present invention. Figs. 3A and 3B show, when superimposed, in block form the apparatus of the executive office. Figures 4A, 4B, 4C, 4D, 4E show, when superimposed, the detailed circuit construction of the executive office apparatus. Figures 5A and 5B, when superimposed, are diagrams of the operation of the executive office apparatus. Figures 6A and 6B provide, when superimposed, a block diagram of the field station apparatus. Figures 7A, 7B, 7C and 7D show, when superimposed, detailed circuit constructions for a typical field station. Fig. 8A and 8B form, when placed

superimposed, a waveform diagram of the operation of the field station apparatus.

Fig. 9 is a diagram of the working method and the circuit connections for a station step shifter that is used at each field station and

fig. 10 shows the use of the field station apparatus for controlling the devices at a typical field station, such as switches and signals.

The designations (B +) and (B —) indicate respectively the positive and negative terminals for a power source with a relatively high voltage. The designation for an earth connection indicates the connection to a voltage level between the (B +) and (B —) voltage levels. The designations (+ ) and (—) indicate the positive and negative terminals for a power source with relatively low voltage.

The code connection system according to this invention can be used in a number of different ways. The particular embodiment described and illustrated here, however, shows the use of the system for the control of switches, signals and other devices at a railway section and the receipt of messages regarding the affected state of these various devices as well as messages about the busyness of tracks and the like.

Fig. 1 shows a pair of line wires 10 extending between a management office CO and several field stations FS. As all the field stations are the same, only a typical field station FS is shown in block form in this diagram.

All the field stations and the management office send and receive over the one pair of line wires 10. The controls are sent over the line wires as a coded carrier frequency Fl. while the indications are sent from each of the field stations to the management office over a different carrier frequency F2. As both carrier wave receivers and transmitters 1 themselves are not sufficiently frequency selective, a bandpass filter 26, 15, 16, 25 has been inserted between the line wires 10 and each of the carrier wave rciottagers 27, 17 and also between the output of each carrier wave transmitter 14, 24 and the line threads 10.

The management office apparatus comprises code control circuits 11 which are controlled by control arms 12. The code circuits 11 comprise means for marking the successive work cycles to create the code which forms each complete cycle. The particular permutation of dash and space characters which form the code group for a particular station is controlled by the position of the control arms 12.

The output of the code control circuits 11 comprises a time-shifted series of code elements therein including the synchronization marking, the various station call markings and the marking and space characters which form the codes for the various field stations. This output is applied to a modulator H3 which controls the outputs of the carrier wave transmitter 14. For each interspace element in the code, a modulator 13 thus prevents the carrier wave transmitter 14 from delivering its particular carrier frequency Fl

through the band pass filter 15 to the line wires 10. For all the other types of elements

which is encompassed by a code control period, the xnodulator 13 causes the carrier wave transmitter to send short pulses of carrier frequency and with specific durations as will be explained later, and these carrier wave pulses with frequency Fl are then applied via bandpass filter et,15 to the line wires 10.

At each field station, the received control code of frequency Fl is printed over et

bandpass filter 16 on a carrier frequency receiver 17. The code output of the receiver 17

is then applied to the code decoding circuits 18.

The code decoding circuits 18 comprise station counters which cause each station to respond only to that part of the control code cycle which is intended for it. As a result of this, a special series of pulses corresponding to the control code for this particular station is printed from the output of the control decoding circuits 18 in the selected station periods. These pulses are applied to a shift register 19 with the result that when the entire control code for the station is received, the various steps of the shift register are successively prepared in accordance with the pattern of

markings and spaces that are determined

for this station. The steps in the displacement register are then able to selectively energize several control relays 20 and these are then able, via their contacts, to control different controlled devices 21.

Prior to the reception of a control code at any particular station, the controlled devices 21 are able, via the indication contacts 22, to selectively prepare the various steps of the shift register 19. The steps of the shift register require a special permutation of what we may call their "0" and "1" states corresponding to the indication code that is to be sent to the manager's office. When the control code circuits 18 are operative to bring the received control code into the shift register 19, the indication code which was previously stored in the shift register is therefore brought out by the last stage of the shift register. The shift register can therefore be used to directly control the transmitter 24 with the result that pulses of the frequency F2 are applied to the line wires 10 above the bandpass filter 25 in a pre-selected combination of markings and spaces which form the desired indication message to be sent to the manager's office.

At the head office, the received indication code is imprinted over the bandpass filter 26 on the carrier receiver 27. As a result, a series of time-separated pulses whose pattern corresponds to the received indication code is imprinted on the indication storage medium 28. The control code circuits 11 operate over the indication gate and the decoding circuits 29 to provide the necessary gate functions with respect to the indication storage matrix 28. This effectively means that the received indication code is routed to the appropriate indication storage circuits associated with the particular field station from which the indication code originates. These indication gate and clearing circuits 29 also act to erase the indications for a particular field station just prior to the reception of a new indication code from that station. In this way, the indications that are continuously displayed represent the latest messages received from the field stations.

The special code structure for the present invention is shown in fig. 2. It is shown in the drawing that each duty cycle begins with a train of two pulses the duration of which is greater than that of the normal marker elements which form either the control or indication code for a particular field station. The first of these long code elements is the synchronization pulse, and the second is the station call pulse for station No. 1. Following the station call pulse for the first field station is a series of time-separated carrier frequency pulses, which pulses by their position permutation in the space between the first and second station call pulses obtains a clear control message for the first field station. After this call pulse for the first field station, the control code includes another station's call pulse and then a series of markings and spaces which provide the message for the control of the devices at the second field station. Similarly, the control code period includes controls for all the field stations 1 the system.

In the present embodiment of the invention, it is assumed that five field stations are arranged, but that each field station receives a group of controls twice within a single working period, so that a larger number of controls and indications can be arranged at each field station without a corresponding increase of the complexity of the equipment. After each of the five field stations has received its group A control codes, they are thus sent to each station in turn and order once more, so that they receive their group B codes. At the end of the second station period for station no.

5, the cycle is complete, and a new cycle will be

immediately started with another synchronization pulse.

During the time when the controls are sent to field station no. 1, an indication code is received from the same field station. Similarly, indication codes are received throughout the cycle from the field stations as they each receive their control codes. Two separate groups of controls are sent to each field station in the system in a complete cycle and at the same time two groups of indication are received from each field station.

A part of the control code period is shown with an enlarged time scale at the top of fig.

2. This makes it clear that the special

carrier frequency Fl used for controls is only transmitted during the first half of each 10 millisecond synchronization pulse period to thereby form the special synchronization pulse. The station call pulse that follows the sync pulse is identical to the sync pulse. In each of the fifteen 5 millisecond pulse periods that occur between the station call pulse for station No. 1 and the station call pulse for station No. 2, a marking or space element is shown in the drawing. There is thus shown a marking denoted by the letter M which is transmitted in the first pulse period and a space denoted by the letter S 1 in the second pulse period. For the marking element, it has been shown that the carrier frequency Fl is only transmitted during approximately the first half of the pulse period. For the spacer element, no carrier wave is transmitted.

The portion of the indication period shown with an enlarged time scale at the bottom of the drawing indicates that fifteen pulse periods are also provided for the transmission of indications from any field station. In any pulse period, either a marker or a gap element can be sent. These mark and space elements are exactly the same as those in the control cycle, except that they are sent on frequency F2 instead. This drawing also indicates that the indication code does not include any synchronization or station call pulses. The reason for this is that the gate functions are provided by the head office, and since the indication code from any field station is received during the same station period during which controls are sent to that station, the head office gate circuits can be directly used to direct the received indication codes to the appropriate indication storage circuits.

The arrangement of the manager's office is shown simplified in fig. 3'A and 3B and in more detail in fig. 4A—4E. In the following description, the working method for the sending of the controls is made clear with the help of the waveform diagrams in fig. 5A and 5B.

The basic timing device in the executive office is the 200 p/s oscillator 35. This Colplbt type oscillator comprises the tube 150 in whose grid-cathode circuit there is a tuned flywheel circuit comprising the self-inductor 151 and the capacitors 152 and 153. The output of this oscillator shown at line A in fig. 5A, is applied over wire 154 to a square wave amplifier 36 comprising tubes 155 and 156. These two tubes are interconnected in such a way as to form a regenerative amplifier type of the kind commonly called a Schmitt trigger circuit. Both tubes 155 and 156 are provided with a common cathode resistor 157. An increase in the voltage on the grid 1 tube 155 causes an increase in conductivity for this tube so that there is a greater voltage loss across its anode resistance 158. The lower anode voltage for this tube is then applied across parallel - the resistor 159 and the capacitor 160 on the tube 156 and causes it to carry a smaller anode current. As the voltage drop across the cathode resistor 157 is mostly affected by the conductivity of the tube 156, there is a smaller voltage drop across the cathode resistor 157 so that the grid cathode voltage for the tube 155 is further increased.

A voltage reduction obtained from the oscillator 35 has a similar cumulative effect in the reverse direction. In this way, any voltage change that occurs on the grid in tube 155 has a greater effect and causes this tube 155 to easily change from conducting with low conductivity to a state with high conductivity. As a result of the large voltage variation that occurs at the anode of tube 155, tube 156 will also easily alternate between full conductivity and a nearly non-conductive state as a sine wave output from oscillator 35 varies between its outer limits. The result is that a voltage waveform appears on the anode of tube 156 which has the approximate shape of a rectangular wave as shown by line B in fig. 5A.

The differentiator 37 comprises the triode tube 161 whose anode is connected across the resistor 162 with (B + ) and whose grid is connected across the resistor 163 also with (B + ). The square wave with output voltage from tube 156 is applied to the grid of tube 161 above the capacitor 164. As the grid 1 tube 161 is directly connected across a large value resistor with (B +), current flows between the cathode and the glitter, which causes the potential on the grid in the tube 161 is reduced by the voltage drop across the resistor 163 to a value which is approximately equal to the voltage for the grounded cathode of this tube.

With each increase in the voltage on the anode of the tube 156, the capacitor 164 is quickly charged over a circuit that includes the low resistance in the grid cathode of the tube 161. The capacitor 164 is therefore charged very quickly and only a very small positive voltage pulse occurs between the grid and cathode of tube 161. Since this tube is completely conducting because its grid is at nearly zero grid-cathode potential, this slightly positive pulse has no appreciable effect on the output of this tube.

When the anode voltage for tube 156 has been at such a high level that capacitor 164 is charged to a relatively high voltage, the sudden voltage drop for tube 166 caused by the negative edge of the square wave will cause capacitor 164 to discharge. As the voltage across capacitor 164 cannot be changed instantaneously, the sudden voltage drop in the anode circuit for tube 164 will cause a corresponding voltage drop between grid and cathode for tube 161. In this way, tube 161 is immediately driven to block so that current can no longer flow from the cathode neither to the grid nor ti] the anode. The anode voltage for this tube will also immediately rise to a high value which is approximately equal to the value at (B +). The capacitor 164 can now only be discharged across the resistor 163 which has a very high resistance value so that the capacitor 164 can only be discharged slowly. Thus, while the capacitor 164 is being discharged, the tube 161 is kept non-conductive so that its anode voltage remains approximately at the (B + ) level. After an interval determined by the time constant in the discharge circuit for the capacitor 164, this capacitor is sufficiently discharged that the grid potential rises above the blocking level, and the tube 161 again becomes conductive. This tube's anode voltage is then rapidly lowered to obtain the trailing edge of the positive output pulse.

The differentiating device 37 works in this way in response to each negative voltage variation that occurs in the square wave output from the square wave amplifier 36. Each of the resulting positive-going pulses (see line C, Fig. 5A) is applied across a capacitor 185 to the grid in the tube 166 which is contained in the pulse delay device 40 and is also applied to a cathode follower 38 above a capacitor 167.

The cathode follower 38 as well as the other cathode followers which are included in the structure of the present invention are of the usual type and are therefore only shown in 1 block form for the sake of simplicity. These cathode followers are normally biased so that they become non-conductive or nearly non-conductive, resulting in a cathode voltage of zero volts, but they become conductive when the grid is driven positive to provide a positive output voltage at the cathode.

The pulse delay device 40 comprises the two tubes 166 and 168 which are each connected so that they work as a differentiating circuit similar to the differentiating device 37 which has just been described. With the trailing edge of the output pulses going in each negative direction, which is provided for by means of the differentiator 37, tube 166 thus becomes non-conductive with the consequence that you get a sudden increase in its anode potential, and this tube is then brought back to its conductive state only at the end of an interval determined by the time constant of the discharge of capacitor 165. When capacitor 165 is sufficiently discharged to allow tube 166 to become conductive again, its anode voltage will rapidly drop to provide the 1 negative-going trailing edge of the positive pulse output from this tube.

The positive-going leading edges of the voltage pulses that appear on the anode of tube 166 have almost no effect on the line 1 tube 168. The negative-going trailing edges, however, cause the capacitor 169 and the blocking tube 168 to discharge so that a sudden increase in the anode voltage occurs on the anode of tube 168. After a predetermined interval, capacitor 169 is sufficiently discharged to allow tube 168 to conduct again so that the anode voltage of this tube 168 is lowered to its normal value. As a result of this, positive-going trigger pulses of relatively short duration appear on the anode of tube 168. These pulses are delayed in relation to those delivered by the differentiating device 37 and the value of this delay corresponds to the length of time in which tube 166 remains non-conductive, as it only is the restoration of the conductivity of this tube which causes the tube 168 to become non-conductive for a time and provides a positive-going output pulse at its anode. These output pulses obtained from tube 168 are applied via a capacitor 170 to a cathode follower 41. The corresponding positive-going trigger pulses obtained from the output of this cathode follower 41 are then applied to the carrier wave modulator 13 shown in fig. 4C. These pulses are only slightly delayed in relation to those delivered by the differentiating device 37 and are shown by line C in fig. 5A, so that they are not shown in this fig. 5A.

The control step mover 43 which is shown in detail in fig. 4A comprises several series-connected bistable trigger circuits. A separate step is provided for each of the 15 steps comprised by a station period with an additional step provided for a station call pulse step and a further step for a station call rest step.

Each step in the control step device 43 has two interconnected triode tubes.

The interconnection is made in such a way that only one tube in the pair can be conductive at any time. The other pipe in the pair is then controlled to a fully blocked state. The conductive state of one tube causes a weak lattice voltage on the other tube, so that such a tube is kept non-conductive. The non-conducting state for such a tube tends to provide a high lattice voltage for the other tube to keep it fully conductive. The relative conduction states of these tubes can be reversed almost instantaneously by applying the correct input voltage to a suitable electrode in one of the tubes.

Considering the step of the control step device 43 corresponding to step No. 1 for a station period, the tube 175 is shown to have its left cathode connected directly to ground, while its right cathode is connected to a pulse transfer terminal 39. The control grid for the right side of the tube

175 is connected across a resistor 177 with

(B —) and is also across a resistance 178

connected to the anode of the left side of the tube 175. The left grid of tube 175 is connected via a resistor 179 to a clamp 180 and is connected via a resistor 181 to the right anode of the tube 175. The right and left anodes of this tube are connected across resistors 182 and 183 respectively to (B +). The right grid for tube 175 receives an input voltage across capacitor 184 and resistor 185 from the right anode of the tube included in the previous stage of the control step device. The voltage that appears on the right anode in tube 175 is also applied across a resistor 186 and capacitor 187

on the following step of the control step device.

Rail 180 is generally connected across a normally closed reset pushbutton 176 shown in FIG. 4B to (B —). In order to properly set the various steps of both the control switches and the station switches, when the system is first powered up, the reset pushbutton 176 is depressed to remove the (B — ) voltage from the rail 180. As a result, no longer the resistor 179 1 the grid circuit for the last half of the tube 175 its lower clamp connected to (B —). However, the left grid of this tube is still connected across resistors 181 and 183 to (B + ) so that a relatively high voltage tends to appear between the left grid and the grounded cathode of tube 175. The left half of tube 175 is therefore made conducting and the resulting voltage drop across its anode load resistance causes its anode voltage to be brought down to a relatively low level.

Between the occurrence of pulses from the output of the cathode follower 38, rail 39 is approximately at ground potential. The right cathode in tube 175 is therefore normally also earthed. The voltage on the right control grid is at a certain value between the relatively low positive voltage on the anode of the conductive left half of the tube and (B —). The actual level of this voltage as determined by the relative value of resistors 177 and 178 is selected to provide a voltage on the right grid sufficiently negative to bias the right half of the tube over blocking.

This condition for the stepper stage where the left half of the tube 175 is conductive and the right half is non-conductive is a steady state and is maintained when the reset push button 176 is returned to its normal closed position to connect rail 180 to ( B —). The right half of the tube 175 is thus kept non-conductive because the low anode voltage for the other half of the tube causes it to have a high negative grid cathode voltage. At the same time, the high anode voltage on the right part of the tube 175 raises the grid cathode voltage of the left half sufficiently to ensure that this half of the tube is fully conductive.

This condition for the control steps in which the left half of the tube is conductive and the right half is non-conductive is referred to as the "0" condition for this step. The opposite state in which the upper-right half is conducting and the left half is blocked is designated the "1" state of the step.

When any stage in the control step device is brought back from its "l" state to the "0" state, a positive voltage variation is applied to the next stage. Thus, with this reversal of step No. <1 the step from the "1" to the "0" state, the right anode will suddenly rise with respect to potential as the anode current ceases across the resistor 183. A positive voltage variation is therefore applied across the resistor 186 and the capacitor 187 on the guide grid for the right half of the tube 188 1 step no. 2 step.

In exactly the same way, the return of the station call step causes a positive voltage variation to be applied across the capacitor 184 on the right grid in the tube 175 which is thus momentarily raised above the blocking potential so that the right half of the tube becomes conductive. The anode current across resistor 183 is momentarily increased with a consequent reduction of the voltage on the right anode and also at the right grid for this tube. The left half of this tube then conducts less so that its anode voltage rises and produces an increase in voltage on the right grid so that the right half conducts more anode current. The phenomenon is thus cumulative and results in an almost instantaneous transfer of relatively conductive states for the two halves of pipe 175.

Step No. 1 the stage, when thus controlled to its "1" state by a positive voltage variation obtained from the previous step, remains in this state until a positive voltage pulse appears on rail 39. The rise in voltage on the right cathodes of tubes 175 results in a sudden lowering of the grid cathode voltage for the right half of this tube so that it immediately conducts less current. The cumulative coupling action just described now works in reverse and immediately causes the right half of the tube to become blocked and the left half to become fully conductive so that the stage is returned to its "0" state. There then again occurs a sudden rise in the voltage on the right anode so that a positive voltage variation is impressed on the right grid in the tube 188 and causes this next stage to immediately work its way to its "l" state. Step no.

2 step remains 1 this state Until a

second occurrence of a positive voltage pulse on rail 39 causes step #2 to return the step once more to its "original" state.

From this description it will be seen that each occurrence of a positive voltage pulse on rail 39 causes any stage in its "l" state to be brought back and at the same time causes such "1" state to be transferred to the next stage.

In order to achieve the desired step operation, the control step device 43 is brought to start by actuation of the reset push button 176 in FIG. 4B, so that only a single stage of the control step device is initially in the "1" state, and all the remaining stages are in their "0" states. The action of the reset push button therefore ensures that step No. 1 is initially in its "0" state. The same situation applies in relation to the remaining stages of the control step device with the exception of No. 14. Pipe 189 in this step No. 14 stage has its right guide grid and not its left guide grid connected to rail 180. The removal of the (B —) voltage from rail 180 therefore ensures that its right half becomes conductive and its left half non-conductive, being its "l" state. The reason for the setting of the control step device so that step No. 14 is the particular step selected to initially be set to its "l" state will become apparent later.

The station step device 44 which is shown in fig. 4B is similar to the steering step device 43 which has just been described. A separate stage is provided in the station stepping device for each station period in the cycle with an additional stage provided for the synchronization pulse step. The left grid of the tube for each stage is connected to rail 180, but the sync stepper stage has the right grid of its tube connected to rail 180. The sync pulse stepper is therefore initially set by actuation of the reset push button 176 to its "1" state, but all the remaining stages of the drive stepper are initially set to their "0" states.

The right cathode of each stage included in the stepper device 44 is connected to a rail 49 which receives a voltage from the cathode follower 191 associated with the stepper trigger 48. Positive trigger pulses are caused to appear on rail 49 in a manner that will be described below, and each such pulse causes the particular stage in the drive stepping device 44 then in its "1" state, in which the right half of its tube is conductive, to be returned to the "0" state in which the right half of the tube is not -leading. This transition of the fit stage to its "n" state then causes a positive voltage variation to be applied to the right grid in the tube for the following stage and causes such next stage to transition to its "l" state.

The station step trigger 48 receives its input from the left anode of the tube included in the step #15 stage of the control step device 43. When the step #15 stage is finally brought from its "1" state to its normal "0" state at a pulse that occurs on rail 39, the left half of tube 195 becomes conductive so that a negative voltage variation occurs on the left anode and is applied across wire 192 and across capacitor 193 on the grid in tube 194 which is included in the station step trigger 48.

The station step trigger 48 is essentially a differentiator similar to the differentiator 37 shown in fig. 4A. Any positive voltage variation occurring on the left anode of tube 195, which occurs when the step #15 stage is transferred to its "1" state, has virtually no effect on the anode output voltage of tube 194. The negative voltage variations which occurs, however, when step #15 the stage is returned to its "0" state, causes tube 194 to be driven into blocking so that its anode voltage suddenly rises and causes a positive voltage variation to be impressed across capacitor 196 at the input of cathode follower 191' whereby a positive trigger pulse is obtained on rail 49 to the various stages of the station step device 44.

From this description it will be seen that every time the control step device 43 has been affected over all its steps and with its last step no. that the station step device 44 is moved forward over a single step.

The one step advance of the station step device 44 causes an input to be applied to the control step start trigger 52 in fig. 4C with the result that a pulse is impressed across wire 53 on the station call pulse step of the control step device 43. This pulse causes the station call pulse step to transition to its "l" state so that the following occurring pulses on wire 39 can produce the desired step operation by transmission of this " l" state to the following steps in the control step device 43.

Alternating steps of the station step device 44 are connected with rail 55 and the remaining alternating steps are connected with rail 56. F.g. are the right anodes of the double triodes included in the synchronizing pulse stage and the group A station No. 2 stage respectively connected across anode resistors 200 and 201 with rail 95 which is then connected across the primary winding of a transformer 202 with (B + )• The right anodes of the double triode tubes comprised by the Group A station No. 1 stage, the Group A station No. 3 stage and the Group A station No. 5 stage (not shown in Fig. 4B) are all connected to rail 56 which is then connected across the primary winding for transformer 203 with (B +).

When any stage of the drive stepping device 44 is brought back to its "0" state from the "1" state with the result that the right half of the corresponding tube becomes non-conductive, a current reduction will occur in the primary winding of the associated transformer which is included in the steering step start trigger 52. F. e.g. forcing the group A station No. 2 stage to its "0" state will cause the right half of tube 204 to become non-conductive so that a current reduction occurs across resistor 201 and through the primary winding of transformer 202. The resulting flux change in the primary winding induces a voltage in the associated secondary winding. This secondary winding is connected in parallel with a resistor 205 whose purpose is to dampen the waveform of the voltage that appears across the secondary winding. The negative voltage pulse that occurs at the upper clamp of the secondary winding is then applied across the resistor 206 on the control grid for the triode tube 207.

Tube 207 is normally conductive because its cathode is grounded, and its control grid is also normally at ground potential. Because of the anode current through its anode load resistor 208, the anode voltage for this tube is normally at a relatively low value. The negative voltage pulse applied to its control grid results in the tube momentarily being driven into blocking so that its anode potential suddenly rises to the approximate level (B +). The capacitor 209 will then be charged and causes the control grid for the cathode follower triode tube 210 to be momentarily made positive. Tube 210 is normally blocked due to the connection between its control grid and (B — ) across resistor 211 so that no voltage drop occurs across cathode resistor 212, and wire 53 is therefore at ground potential. The charging of the capacitor 209 and the consequent increase in the grid cathode potential of this tube 210, however, causes the tube to momentarily conduct so that a positive voltage appears at the cathode and across the wire 53 is applied directly to the left cathode of the tube 262 which is included in the station call pulse step. for the steering step device 43 (see Fig. 4A).

When a step of the station step device 44 which is connected to rail 56 instead of rail 55 is influenced to its

"0" state the effect is the same as

just described, with the exception that a current variation then takes place in the primary winding for transformer 203 instead of in transformer 202. Tube 207 then receives a negative voltage pulse on its control grid from the secondary winding for this transformer 203. This pulse acts in the same way to produce a positive pulse on wire 53 to transfer the station call pulse step of this control step 43 to its "l" state. Successive stages of this station step device 44 must be connected across separate rails to separate transformers for the control step start trigger 52, because a reduction of the current occurring in one stage when this stage is actuated to its "0" position is accompanied by an increase of the current to the right half of the pipe of the previous stage when that stage is transferred to its "l" state. If such two successive stages were connected to the same transformer winding, there would essentially be no current change in such a winding, because the increase in current in one tube would be counteracted by a corresponding current reduction in the other tube.

From the foregoing description it will be seen that it is clear that a continuous stepping operation occurs in relation to both control and station step devices as long as a trigger pulse is applied to rail 39. Each complete operation of the control step device will thus occur when step no. 16 is transferred to its "0" state upon the application of a trigger pulse on rail 49 so that the station step device 44 will advance one step. With this step advancement at the station stepper 44, a trigger pulse is generated by the control step start trigger 52 to transfer the first step of the control stepper 43 to its "l" state. In this way, it is possible for pulses appearing on wire 39 to again effect a step operation of the control step device. This chain of events occurs again and again and thereby marks the station periods by the action of the station step device and the different pulse periods included in each station period by the action of the steering step device.

At the end of a cycle, a variation occurs with regard to the just described repetition operation of the control and station step devices and the purpose of this variation is to insert at the beginning of the new cycle the required synchronization pulse. Accordingly, when Group B drive #5 drive stepper 44 is returned to its "0" state at the end of a cycle, an output will be applied to step #14 trigger 67 (see lines L and P of FIG. 5A ). This is in contrast to the previously described operation in which the return of any step of the drive step device to its "0" state will cause an output to be applied to the control step start trigger 52.

This control which the step no. 14 trigger receives over wire 66 operates in much the same way as the corresponding controls applied to the control step start trigger 52 from any other step in the station step device. The fact that the right half of pipe 211 in group B station no. 5 step is made non-conductive thus causes a negative voltage variation to be applied to the control grid for pipe 213 which is included in this step no. 14 triggers 67. The normally non-conductive cathode-follower tube 2)14 momentarily becomes conductive so that a positive voltage pulse is applied across wire 68 to the left cathode in step No. 14 stage of control step device 43. This step No. 14 stage is at this time in its normal "0" state >with the left half of its tube 189 conducting. The positive voltage pulse applied to the left cathode causes the left half of the tube to become non-conductive and the right half conductive so that this stage is actually transferred by this voltage pulse to its "l" state (see line I, fig. 5A).

The next positive trigger pulse occurs on wire 39 five milliseconds later and causes the step #14 stage to return to its "0" state and the step #15 stage to then assume the "1" state (line J, Fig. 5A). Five milliseconds later, another trigger pulse on rail 39 causes step #15 the step to return to its "0" state, and a pulse is then applied across , wire 192 of station step trigger 48 (line

K).

The return of the group B station No. 5 stage for the station trigger device 44 not only produces the results just described in relation to the control step device 43 but also causes: that a trigger pulse is applied across wire 57 i to the synchronization pulse step stage for the station step device (see line N) . The increase in potential which occurs at the anode of the right half of tube 211 when this final stage of the station stepping device is returned to its "0" state causes a positive voltage variation to be impressed across resistor 215 and capacitor 216 on the right grid of tube 217 in the synchronization pulse step. The positive pulse which then occurs between the grid and the cathode on the right half of the tube 217 reverses the state for this stage so that the right half becomes conductive and the left half non-conductive. Consequently, at the time when the step No. 15 steps of the control stepper 43 are affected after the end of the cycle, the synchronization pulse stepper of the drive stepper 44 will be in its state "1".

Two pulse periods later, step #15 is returned to its "0" state (see line J, Fig. 5A) with the result that the station step trigger 48 responds by causing a pulse to appear on wire 49 (see line K). This pulse causes the synchronization stepper for the station stepper 44 to return to its "0" state (see line M) and the group A station No. 1 step is brought to its "1" state (see line N). It can thus be seen that the sync pulse stepper only remains in its "1" state for two pulse periods totaling ten milliseconds. All other steps in the drive step device 44 of course remain in their "l" states successively for a period of time required for the control step device 43 to function through all its 17 steps. It will later be described in detail in connection with a consideration of the carrier modulator 13 that the control carrier frequency is transmitted when both the synchronization pulse step and the step No. 14 step are simultaneously in their "l" states. The carrier wave is thus only transmitted in the first five milliseconds of a total of ten milliseconds that the synchronization step is in its "1" state and thereby provides the necessary synchronization pulse at the beginning of each cycle in the manner shown in fig. 2.

In each step of the control step device 43 except the station call pulse step is

its left anode connected to the input gate circuit of a cathode follower amplifier. When any stage is in its zero state with the left half of the tube fully conducting, the anode voltage for the left anode is at a relatively low value so that a correspondingly low voltage level is applied to the grid circuit of the associated cathode follower. This cathode follower

is 'biased so that the low voltage it receives from the corresponding control step under these conditions results in the cathode follower tube being blocked. Now, however, any stage of the drive step device is in the "l" state where the left half of the tube is non-conductive, the anode voltage applied to the grid circuit of the associated cathode follower is of a considerably higher value so that the cathode follower tube becomes conductive and delivers an output voltage 'with a relatively high value.

The cathode followers connected to steps Nos. 1 to 15 of the control step device 43 control the voltages appearing on rails 22.0 to 224. Although only five such rails are shown in FIG. 4A and 4B, one for each of the five steps shown by the control step device, there is actually a rail provided for each of the 16 steps corresponding to the 15 pulse periods in each station period.

The matrix output cathode followers 62 shown in FIG. 4B, comprises multiple dual gate cathode follower amplifier stages with one stage for each station period of a duty cycle. The anode for each of these cathode followers 62 is selectively connected over the control arms 61 with the rails 220 to 224. The control arm 225 e.g. is a three-position coupler that will selectively connect the anode of the cathode follower tube 226 to one of the rails 222, 223 or 224. The two-position coupler 227 functions in a similar way to selectively connect the anode of tube 226 to either rail 220 or 221. Additional copies that are not shown in fig. 4B can be arranged so that the anode for tube 226 can be selectively connected to the remaining of the 15 rails comprised by the control matrix 60.

Each stage of the drive stepper 44 with the exception of the sync pulse stepper also has an associated cathode follower stage which provides an approximately zero cathode output voltage when the respective drive stepper is in the "0" state, but provides a relatively high output voltage when such a stage is in its "1" state. -state. Each of these cathode followers delivers its output across a resistor to the control grid in a corresponding one of the matrix output cathode followers 62. The output voltage obtained at the cathode in the cathode follower 228, for example, is applied across the resistor 229 on the control grid in the cathode follower tube 226.

In order for any matrix-1 output cathode follower to become conductive, both its anode and control grids must simultaneously be impressed with positive gate voltages. As each stage of the drive step device 44 is in turn transferred to its "1" state over a duty cycle to thereby mark a drive period, the respective

'matrix output cathode followers' control grids supplied with positive gate voltage. During the group A station's No. 1 period, the cathode follower 226 has its grid opened and during the group A station's No. 2 station period, the cathode follower tube 230 has its control grid supplied with positive gate voltage.

Each matrix output cathode follower 62 has its anode positive gate voltage supplied for certain pulse periods during the station period in which its grid is then supplied with gate voltage. During Group A station's No. 1 period, e.g. positive gate voltages appear every five milliseconds duration successively on rails 220 to 224. During step no. 1 pulse period a positive gate voltage appears on rail 220 but it cannot be applied to the anode in tube 226 if the coupler 227 is in its left position as shown on fig. 4B. During step no. 2 the pulse period, a positive gate voltage appears on rail 2211 and it is applied across the closed contact of coupler 227 and the separating rectifier 231 to the anode of the cathode follower tube 226. In a similar way, positive gate voltage occurs on steps no. 13 to 15 during the pulse periods gate voltages successively on the rails 222 to 224. With the three-position coupler 225 in its middle position, however, only the gate voltages that appear on rail 223 can be applied to the closed contact for this coupler 225 and across the separating rectifier 232 on the anode for tube 226.

In these pulse periods during Group A station No. 1 station period, during which the anode for. tube 226 is supplied with gate voltage as selected by the positions of the control arms, the matrix output cathode follower tube 226 becomes highly conductive so that a positive voltage is produced across resistor 233 which is the common cathode resistance for all these cathode followers. This positive voltage is then applied via wire 64 to the control grid for pipe 234 which is contained in the carrier wave modulator 13 in fig. 4C.

In a manner similar to that just described, each matrix output cathode follower tube in the particular station period where its control grid is supplied with positive gate voltage can become conductive in certain selected pulse periods for this station period in accordance with the positions of the control arms associated with this particular cathode follower . Since the positive pulses which are thus caused to appear on wire 64 and which are impressed on the 'carrier wave modulator L3 result in the transmission of the carrier wave so as to produce markings for the control code, it can be seen that the choice of the pulse periods in which markings are to be sent in which any station period, is completely in accordance with the positions of the control arms corresponding to this station period. In this way, a two-position device on a field station such as a switching machine can be controlled using a two-position control arm, such as the control arm 227. In one position, the control arm causes a marking pulse to occur for the code in the first pulse period and this can then be decoded by the field station representing "switching normal" control for the switching machine. With the control arm in the other position that a marking is sent in the second pulse period, the resulting code can be decoded at the corresponding field station and translated into a determination of a coupling control in the reverse direction for the relevant coupling machine.

A more complex control can be carried out at a field station by using three successive pulse periods with a three-position control arm which determines in which of the three pulse periods a marking character is to be sent. A marking in the first of the three pulse periods can e.g. indicate a clearance control for a signal for eastbound traffic, a marking in the second pulse period can then indicate a stop control for the signal, and a marking in the third pulse can then indicate a clearance control for the signal for westbound traffic.

The purpose of the carrier wave modulator li3 which is shown in fig. 4C, is to open the carrier wave transmitter 14 at appropriate times during a duty cycle so that the various control code signals including station call markings and synchronization markings are sent to the various field stations.

The carrier wave modulator receives an input across the wire 64 from the matrix output cathode followers 63. As already described, the voltage on this rail 64 is normally approximately at ground potential but is raised to a positive value during each pulse period where a check mark is to appear in the control code. When the wire 64 is at its normal ground potential, the tube 234 of the carrier modulator 13 is non-conductive due to the negative bias potential impressed on its control grid from (B — ) across the resistor 240. Since there is no anode current drop across the anode load resistor 241 for this tube, a high voltage appears on the anode of tube 234. This voltage is applied across resistance 242 to the control grid of tube 243 and is sufficiently high to overcome the negative

bias that is impressed on its lattice from

(B —) over resistance 244. Tube 243 becomes føl-. can be made conductive only provided that the required high voltage is applied to its anode across the resistor 245.

The voltage for the anode on tube 243 is obtained from the pulse delay cathode follower 41 in fig. 4A across wire 42. Each positive trigger pulse obtained from this cathode follower 41, which occurs at a time when the grid of tube 243 is raised so that this tube conducts, encounters a low impedance circuit between the anode of tube 243 and ground, so that such positive trigger pulses in reality are short-circuited. Each pulse that is short-circuited in this way has almost no effect on the capacitor 246 on the control grid in tube 247.

However, when a marker is to appear in the control code, the voltage on rail 64 is raised so that tube 234 becomes conductive. The anode voltage for this tube is then lowered due to the voltage drop across resistor 241. This lower anode voltage causes tube 243 to be biased above blocking. Accordingly, any positive trigger pulse applied across resistor 245 to the anode of tube 243 at such a time will encounter a high impedance between this anode and ground so that the pulse is not short-circuited. Each such trigger pulse, when pressed over the capacitor 246 on the grid in tube 247, becomes effective so that the grid cathode potential on tube 247 is immediately raised.

The trigger pulses which cause stepping operations for the hard control and station stepping devices are obtained from cathode follower 38 which is connected to differentiator 37. The trigger pulses which are applied to the anode of tube 243 across the resistor 245 are, however, obtained from the output of cathode follower 41 which is connected to the pulse delay 40. These trigger pulses are only very slightly delayed compared to those obtained from the cathode follower 38. The reason for this delay is that the gate pulse that is applied to the grid in tube 234 has its working time and time for removal approximately at the same time as the pulses that appear on rail 39 when control and the station chalk devices take their respective steps in response to these pulses on rail 39. The small delay provided for the pulses applied to tube 243 ensures that these pulses occur after all required switching actions have taken place.

(Tubes 247 and 248 are included in a circuit commonly referred to as a "one-shot" multivibrator. An approximately zero grid-cathode potential is provided for tube 248 by connecting its control grid across resistor 249 to (B+). The leading

condition of this tube 248 causes it to have a relatively low anode voltage due to the voltage drop across the resistor 250. This voltage, when applied across the resistor 251 on the control grid for tube 247, will therefore not be of sufficient strength to overcome the negative bias that is arranged across the resistor 252 from (B—) so that the tube 247 is normally non-conductive.

When a positive trigger pulse is applied across the capacitor 246 to the control grid in tube 247, the grid cathode potential for this tube is immediately raised above the blocking level so that this tube begins to conduct. The capacitor 253 which is normally charged to a relatively high voltage is then discharged across the conducting tube 247 with the result that the grid in the tube 248 is suddenly made negative in relation to its grounded cathode. Tube 248 thus becomes non-conductive, and the high anode voltage which it then provides across resistance 251 to the control grid of tube 247 causes tube 247 to remain conductive even after the termination of the positive pulse on its control grid. The reversal of the conducting states of tubes 247 and 248 continues as capacitor 263 discharges across tube 247 and across the relatively high resistance 249. However, after an interval whose length is determined by the time constant for the discharge of capacitor 253, the grid potential of tube 248 rises to a level which allows this tube to become conductive again. A similar cumulative switching action then takes place which reverses the conduction states of these two tubes, bringing them back to their original states practically instantaneously.

The multivibrator comprising tubes 247 and 248 is arranged to remain in its abnormal state in response to positive pulses applied to the grid in tube 247 for a time approximately equal to one half of a five millisecond pulse period. At the beginning of each pulse period in which a mark is to be sent, tube 248 therefore becomes non-conductive and tube 247 conductive, but this condition is maintained only for the first half of the pulse period. At about the middle of the pulse period, the multivibrator returns itself to its normal state so that tube 248 again becomes conductive and tube 247 is blocked.

In the normal state of this multivibrator circuit, the anode voltage for tube 248 is at a relatively low level due to its conducting state. This voltage is applied across the resistor 254 to the control grid in tube 255, but is not sufficient to overcome the negative bias applied to this grid from (B—) across the resistor 256. The tube 255 thus normally becomes non-conductive so that no anode current flows in this tube across resistor 257. At the beginning of any pulse period in which a marker is to be sent, tube 248 becomes non-conductive so that its anode voltage rises and causes tube 256 to conduct. There is then a significant anode current in this tube 255 across resistance 257 so that a negative voltage variation is applied at this time across the capacitor 258 to the control grid in tube 259. When the multivibrator circuit is restored to its original state, tube 255 again becomes non-conductive so that the anode voltage suddenly rises and the voltage on the grid in tube 259 is then raised to its normal value.

Tube 259 is normally in a fully conductive state due to the connection of its control grid across resistor 260 with (B + ). The negative voltage variation that occurs at the beginning of each marker causes capacitor 258 to discharge, thereby suddenly lowering the grid cathode voltage on tube 259 so that this tube becomes non-conductive. Since the carrier wave transmitter 14 is operative so that it delivers its specific carrier frequency Fl to the bandpass filter 15 every time tube 259 is controlled so that it becomes non-conductive, it is clear that this carrier frequency will be sent in each pulse period in which a marking character is desired to be sent. It further follows that since the multivibrator comprising tubes 247 and 248 only remains in its abnormal state for the first half of each such pulse period, the carrier frequency will likewise only be transmitted for the first half of each pulse period in accordance with the general coding scheme shown on fig. 2.

Tube 261 also has its anode connected across the same anode load resistor 257 with (B+). The control grid of this tube is connected across resistor 529 to the station call pulse stepper for control stepper 43. When this station call pulse stepper is in its normal or "0" state, its left anode is at a low potential, because the left half of tube 262 in this step is then leading. This low anode voltage causes tube 261 to be normally non-conductive, as a negative bias is applied to the grid of this tube from (B—) across resistor 263. However, whenever the station call pulse stepper for the control stepper is in its normal "1" state, it rises left anode in tube 262 with respect to potential and causes an increase in the voltage on the grid in tube 261 throughout the following pulse period. The control grid in tube 261 is then raised sufficiently in potential so that this tube becomes conductive with the result that a negative voltage variation is impressed across the capacitor 258 on the control grid in tube 259. The carrier frequency Fl will consequently be sent over the entire five millisecond period that the station call pulse stepping device is in its "1" state. This constitutes the first half of each station call pulse. There is no such connection from the station-call rest cycle to the carrier modulator 13 so that the carrier frequency Fl cannot be transmitted at the next pulse period. In this way, the second half of each station call pulse is formed. It is thus ensured that at the beginning of each station period there shall be a five millisecond pulse period in which the carrier wave is sent followed by a five millisecond period in which no carrier wave is sent to thereby form the special station call pulse. Each of the following pulse periods at steps 1 to 15 is of the same duration, but in each of these the carrier frequency can only be transmitted in the first half of such pulse periods as determined by the working time of the multivibrator included in the carrier wave modulator 13 .

The distinct synchronizing pulses which occur at the beginning of each cycle are provided by the control of tube 266 which is included in the carrier modulator 13. Tube 266 is essentially a cathode follower which receives its anode voltage from the cathode follower 72 which is connected to step No. 14 the step of the control step device 43 (see Fig. 4A). The anode of this cathode follower tube 266 thus receives the anode voltage required for it to become conductive only in the fourteenth pulse period of each station period. The control grid in this tube 266 has a positive gate voltage which is selectively applied to it across wire 73 from the cathode follower 71 which is connected to the sync pulse stepper for the station stepper 44. This control grid therefore receives a positive voltage which causes it to conduct only when the sync pulse step is in its "l" state. The synchronization pulse step is in its "1" state only for the two pulse periods required to go through steps 14 and 15 for the control step device 43. Tube 266 therefore receives the two required gate pulses on its anode and control grid simultaneously only during the first of the two pulse periods in which the sync pulse stepper is in its "l" state. When tube 266 is gate controlled in this way, it becomes conductive so that its cathode, which is normally at ground potential, is raised in voltage as a result of the anode current across the cathode resistor 304. This increase in cathode voltage is applied to the control grid in tube 305 and causes this normally blocked pipes become conductive. In this way, the grid in tube 259 receives a five millisecond negative pulse which causes it to be blocked. As a result, the carrier frequency F1 is again sent in such a pulse period to obtain the first half of the signal pulse in accordance with the general coding arrangement shown in fig. 2.

The carrier wave transmitter 14 comprises tubes 259 and '264 which are connected in a circuit known as a Hartley oscillator. The output of this oscillator is applied to the grid in a power amplifier tube 265. The frequency-determining resonant circuit for the oscillator comprises capacitors 267 and self-induction 268 which are connected in parallel in the cathode circuit for tube 259. When the control grid in tube 259 is at such a voltage that this tube is conductive, the oscillator inactive. However, when the grid in tube 259 is driven into blocking or past with respect to the cathode, the oscillator becomes operative with the result that a sine wave voltage of the desired frequency appears on the cathode of tube 264. The detailed operation of this oscillator will not be described here as its construction and working methods are well known. Furthermore, various other pulse-driven oscillators can also be used. The potentiometer 269 connected between the cathode of tube 264 and ground provides a means of selecting the desired level of input to the grid in tube 265. The tube 265 is cathode biased as a result of the anode current passing across its cathode resistor 236. This resistor 236 connected in parallel is a capacitor 237 whose purpose is to avoid degeneration of the frequency to be amplified. The anode alternating current in the tube resulting from the corresponding alternating current voltage applied to its control grid induces a voltage in the secondary winding of transformer 238. The primary winding in this transformer is included in the anode circuit for tube 265. The secondary voltage is applied across the bandpass filter 15 to the line wires 10. It can thus be said in sum that the transmitter is normally inactive so it does not deliver the determined carrier frequency to the line wires 10, but becomes active, so it does, during each interval when the grid in tube 259 is driven into blocking.

The layout of a field station is shown in simplified form in fig. 6A and 6B and in more detail in fig. 7A—7D. In the following description, the operation of the waveform diagrams in fig. 8A and 8B.

The control code received at each field station from the management office as an interrupted carrier frequency Fl (Fig. 8A, line A) is applied from the line wires 10 over the bandpass filter 16 to the carrier wave receiver 17. This carrier wave receiver 17 comprises a detector stage so that the output of the carrier wave receiver provides a DC output voltage at a low level when the carrier is received and a higher level DC voltage when no carrier is received. This output voltage from the carrier wave receiver 17 is applied via a capacitor 270 to the control grid in the tube 271 which is comprised by the carrier square wave amplifier 77.

Pipe 271 is preferably of the black type, e.g. 6BN6, which has the characteristic feature that a small variation in its control grid voltage causes the tube to vary from full conductivity to full blocking. Thus, when no carrier frequency Fl is received at a field station, the control grid voltage for tube 271 is at a high level and the tube is then conductive so that its anode voltage is low. However, when carrier frequency F1 is received at a field station, the output voltage of the carrier receiver is at a low level so that tube 271 is blocked and its anode voltage is high. The effect of this is that the anode voltage in tube 271 provides a positive pulse for each pulse of carrier frequency Fl that is received at the field station and the duration of such a positive pulse is equal to the duration of the received carrier frequency as shown in fig. 8A by line

B.

Each such positive pulse is applied across capacitor 272, isolation diode 273 and resistor 274 to the control grid in tube 275 which is included in marker multivibrator 78. Tubes 275 and 276 are interconnected to form a "one-shot" multivibrator similar to that previously described in connection with the executive office app rates. Pipe 276 is normally fully conductive and pipe 275 non-conductive. At the positive leading edge of each pulse obtained from the anode in tube 271, the state of the multivibrator is suddenly reversed so that tube 276 is blocked while tube 275 becomes fully conductive. The length of time that the marking multivibrator remains in this state is chosen to be greater than the 2-y2 millisecond interval during which the carrier frequency Fl is received in a marking control code but less than the 5 millisecond interval through which the carrier frequency is received to form the specified synchronization and station call pulses.

During the time that the marking multivibrator 78 is in its abnormal state with tube 275 conducting, the anode voltage of the non-conducting tube 276 is high (see Fig. 8A, line C). When the marker multivibrator 78 returns itself to its normal state, the anode voltage becomes tube

276 suddenly reduced. The anode voltage for tube 276 is applied across the capacitor 277 to the control grid in tube 278 which is comprised by the delayed marking amplifier 79. This amplifier 79 comprises a differentiator circuit similar to the one previously described, so that the negative voltage variation which is applied to the grid in tube 278 when the marking multivibrator is fed back to its normal state causes tube 278 to become non-conductive for a short time interval depending on the length of time required for capacitor 277 to discharge to its new stable value. The delayed selection amplifier therefore provides, as indicated by line D in fig. 8A, a positive trigger pulse whose leading edge substantially coincides with the trailing edge of the positive output pulse obtained from the anode of tube 276.

Each of these positive trigger pulses obtained from the delayed marker amplifier 79 is applied across the capacitor 279 to the control grid in tube 280 which is comprised by the synchronizing and station call pulse separator 80. This tube 280 is a pen-tode whose screen grid gets its proper working voltage at a connection across resistor 281 with (B+). Both control and catch grids for this pipe are normally negatively biased to prevent conduction in the pipe by connections arranged across resistors 282 and 283 to the clamp (B—). Only by the simultaneous occurrence of positive gate voltages on these control and capture grids can tube 1280 be made conductive.

The control grid is momentarily made positive to make it conductive in response to each positive trigger pulse from the delayed selection amplifier 79. The trap grid, on the other hand, receives a positive gate voltage across capacitor 284 from the anode of tube 271 in the square wave amplifier 77. As will see by comparing the lines D and B in fig. 8A, the required gate voltages for tube 280 occur only at times when there is either a sync or a station call pulse. The square wave amplifier is still operational so it provides a high output voltage to open the trap until the time a positive trigger pulse appears from the delayed marker amplifier only for each sync or station call pulse. For the shorter marking characters appearing in the control code, the output voltage of the square wave amplifier has already been reduced to its normal low value at the time when the associated trigger pulse appears from the delayed marking amplifier. As a result, tube 280 becomes momentarily conductive and operative to provide a negative trigger pulse at its anode only for either a sync or a station call pulse as indicated by line E of FIG. 8A.

Each negative trigger pulse from the anode of tube 280 is applied across capacitor 285 and resistor 286 to the control grid of tube 287 which, together with tube 288, is included in the sync and station call pulse multivibrator 81. This multivibrator works in the same way as the marker multivibrator 78. Tube 288 is normally conductive and pipe 287 non-conductive. With the positive trailing edge of each pulse occurring on the anode of tube 280, the conduction state of these two tubes 287 and 288 is suddenly reversed. With tube 288 non-conducting, a positive voltage variation occurs on the anode of this tube. The return time for this multivibrator 81 is chosen so that it gets a value so that a positive voltage pulse is obtained at the anode of tube 288 as indicated by line F in fig. 8A and which has sufficient duration to provide the necessary energy for an integration circuit associated with the synchronization pulse separator 82,

The synchronizing pulse separator 82 comprises the pentode tube 289 which has a screen grid impressed at the correct working potential by being connected across the resistor 290 with (B+). Both the guide and catch grid for the pipe are negatively biased to such an extent that conduction in the pipe is prevented by connections arranged with a negative source of bias. The control grid is connected across resistors 291 and 292 with (B—) and the catch grid is connected across resistors 293 and 294 with (B—). In order for this tube to become conductive, it is required that both its control grid and trap grid are simultaneously supplied with gate voltages that are positive and of sufficient amplitude to overcome the normal blocking bias that is provided.

The positive voltage pulse obtained from the anode for tube 288 and applied across the capacitor 295 and the resistor 293 on the trap grid for tube 289 raises the potential in this negatively biased grid sufficiently to make the tube conductive. Each pulse 1 obtained from the anode of tube 288 is also delivered via an integrating circuit to the control grid of tube 289. This integrating circuit includes the capacitors 296, 297 and 298 and also the resistor 299. Each individual positive pulse received from the anode of tube 288 causes charging of the capacitors 296 and 207 so that the control grid voltage for tube 289 is raised somewhat in voltage but not sufficiently to restore the tube to become conductive.

When the station call pulses are received at a field station with the usual station period interval between successive pulses, the interval between the application of pulses across this integrating circuit on the control grid in tube 289 is sufficiently large that the charge built up in the capacitors 296 and 297 of a pulse becomes approximately consumed by the discharge of these capacitors before the arrival of the next such pulse. At the beginning of a duty cycle, when the synchronizing pulse is immediately followed by the station call pulse for station #1, two positive pulses are applied in quick succession from the anode of tube 288 across the integrating circuit on the control grid of tube 289. The first of these pulses causes the integrating capacitors to become partially loaded. When the second of these pulses occurs, the still charged integration capacitors are further charged so that the control grid voltage for tube 289 thereby rises sufficiently in relation to the cathode to make this tube conductive. As the trap gate is opened at each of these positive pulses, tube 289 is properly opened at such a time so that it becomes conductive. The instantaneous conduction of this tube which follows from this provides a voltage drop across its anode load resistor 300 so that a negative voltage pulse is impressed across wire 301 and across capacitor 302 on the anode of the gas discharge tube 303 which is comprised by the station step synchronization control 83'.

The negative output pulses obtained from the sync and station call pulse separator 80, which is shown at line E in FIG. 8A, is applied over wire 85 over capacitor 308 on the station step device 86. These pulses are applied directly to the control electrode 209 and are also applied over the pulse delay device 87 on a similar control electrode 310.

The pulse delay device 87 comprises resistors 311 and 312 as well as the capacitor 313. In its normal state, the capacitor 313 is completely discharged so that the control electrode 310 is at ground potential. When 5n negative pulse occurs on wire 85, the capacitor 313 is charged negatively so that the potential on the control electrode 310 decreases. The voltage across the capacitor 313 cannot be changed immediately so that the control electrode 310 reaches its maximum negative potential some time after the other control electrode 309 has reached its maximum negative potential. In response to these pulses, glow can be transferred from any cathode to a neighboring cathode in a prescribed direction, ie from any cathode to the next lower numbered cathode. These cathodes are arranged in a circular pattern around the anode of the tube so that the cathode K10 is actually adjacent to cathode Kl. A pulse that occurs on rail 85, when the glow is on cathode Kl, thus causes the glow to be transferred directly to cathode K10.

The general constructive arrangement of the multicathode tube 314 which is encompassed by the station stepping device 86 and also its general mode of operation termed "Polycathode Counter Tube Applications", is described in an article in the November 1953 issue of the magazine "Electronics".

When the glow is on a particular cathode in the tube 314, there is a voltage drop across the resistance which is connected in series with this cathode. Since the glow is transferred from one cathode to the horse only in response to the station call pulse, the glow remains on any cathode and causes the voltage on that cathode to reach a high level for a time the duration of which corresponds to a single station period. In the different station periods, different system functions must be performed. This is carried out by applying the positive voltage that appears on selected cathodes in the tube 314 and applying these voltages as gate voltages to special circuit arrangements that realize the desired.

As explained earlier, it is necessary that the station step device at each station has annealing on its tube 314 at the correct cathode at the beginning of the period. In order to accomplish this, connections are arranged from two nobokathodes in the tube 314 to the station step synchronization control 83. The connections shown in fig. 7A, are those that would be arranged at field station no. 2. As indicated in fig. 9, at this field station, annealing of the cathode K2 is required for the duration of the synchronization pulse period and Group A Station No. 1 station period for one cycle. For this reason, a connection is provided from the cathode K3 in tube 314 across the separating rectifier 315, capacitor 316 and resistor 317 to the grid for tube 318. A connection is also provided from the anode output circuit for tube 303 to the cathode K2.

Fig. 9 shows the situation that at station no. 2 the annealing is normal on cathode K3 at the end of each cycle. The next pulse which appears on the wire 85 and is applied to tube 314, thus corresponds to the synchronizing pulse and causes the glow to switch from cathode K3 to cathode K2.

During the station period when the cathode glow is on cathode K3, the rise in potential on cathode K3 produces a corresponding rise in grid voltage on tube 318. This tube is normally biased to blocking by connection from its control grid across resistors 317 and 319 to (B-). However, when the glow is on cathode K3, tube 318 conducts so that the anode voltage decreases and causes a negative pulse to appear on the grid of gas discharge tube 303. This negative voltage pulse can have no effect on the wire in tube 303 as this tube is already biased to non-conductive condition by a connection arranged from its grid across the resistor 326 to (B-).

When the occurrence of the synchronizing pulse at the beginning of the next cycle causes the glow to be transferred from cathode K3 to cathode K2 there is a sudden drop in voltage across cathode K3 which causes tube 318 to momentarily become non-conductive. Its anode voltage then suddenly rises so that a positive voltage pulse appears on the grid in tube 303 and causes this tube to ignite.

The anode in tube 303 is normally at earth potential as it is connected directly to earth via resistance 322. The high anode-cathode voltage required for the tube to ignite is obtained by connecting the cathode

for this cross resistor 323 with (B-).

As a result, when the pipe 303 becomes

conductive, its normally grounded anode which is connected to the cathode K2, approximately negative with respect to ground. The cathode K2 then assumes a sufficiently negative voltage in relation to all the other cathodes for this tube to ensure that the line will only take place between this cathode K2 and the anode. It is ensured that under normal working conditions the occurrence of a synchronizing pulse at the beginning of each cycle will result in the location of the annealing at field station no. 2 on cathode K2 and not at any other cathode in the tube.

The first station call pulse immediately following the synchronizing pulse produces a negative voltage variation on the anode of tube 289 in the synchronizing separator 82. This negative pulse impressed across capacitor 302 on the anode of tube 303 causes this tube to be extinguished. As a result, there is no longer any anode current across the resistor 322 associated with the cathode K2. The cathode K2 is then no longer held at a potential below the ground potential but is instead allowed to rise to a positive potential which is determined by the cathode current through the resistor 322.

If the station step device 86 at any field station at any time goes out of synchronism, it is ensured that it is quickly brought back into a synchronized state. At any field station operating out of synchronism, the step work of the station step device 85 possibly causes a positive voltage to be applied to the station step synchronization control 83. This causes tube 303 to be ignited so that the glow is then maintained on a special cathode in the tube as had to be selected at the connections between this synchronization control 83 and the station step device 86. At field station no. 2, which is shown in fig. 7A, will e.g. the influence of the station stepping device 86, when it responds to trigger pulses received from the synchronization and station call pulse separator 80, possibly causing the cathode glow to be transferred from cathode K3 to cathode K2. This causes tube 303 in the station-step synchronization control 83 to be lit as already described. The ignition of tube 303 lowers the potential of the cathode K2 so much that the glow is ensured on this cathode and not on any other cathode in the tube.

Tube 303 remains lit and keeps the glow continuously on cathode K2 until the output pulse is obtained from the synchronizing pulse separator 82 at the beginning of a new duty cycle. With tube 303 thus extinguished, the potential on the cathode K2 can assume the positive potential it usually has, when the annealing is on this cathode. In this way, the station stepping device 86 begins the new cycle with the annealing of cathode K2 as required.

The cathode on which the annealing occurs during the first station period of a cycle also has the annealing during the sync pulse period of the cycle. During the synchronization pulse period, however, this cathode has maintained its potential below the ground potential due to the conduction in the gas discharge tube 303. The negative potential on the cathode of the station stepper tube cannot produce the required gate voltage so that any system function cannot be performed before the tube 303 is extinguished by means of the output pulse from the synchronization pulse separator 82. The system function can thus be carried out only during the first station period of the cycle and not in the second synchronization pulse period.

During the two station periods during which controls are received at any field station, the multi-cathode glow tube of the station stepper 86 glows on either cathode K1 or K6, depending on whether Group A or Group B controls and indications are being transmitted. In both of these station periods, the positive voltages appearing on the cathodes K6 and K1 are applied across capacitor 330 and wire 331 and resistor 332 to the control grid for tube 333 which is included in the oscillator gate control 95. Tube 333 is normally biased to blocking due to the connection which is arranged for its control grid across the resistor 334 to (B-). When this tube is in this normal non-conductive state, it has a relatively high anode voltage, because there is then no voltage drop across the anode load resistor 335. However, when a positive gate pulse is applied to the control grid in tube 333 during selected station periods, this tube becomes conductive as that its anode voltage suddenly decreases and remains at such a low value throughout the station period.

The 200 p/s oscillator 96 is of the type known as a pulse driven Colpitt's oscillator. This oscillator encompasses the two

triodes 336 and 337, the tube 336 having a

frequency-determining resonant circuit in its cathode circuit. This resonant circuit includes

the self-inductance 338 and the capacitors 339 and 340. The grid in tube 336 normally seeks to be held at ground potential as it is connected via resistance 341 to ground. Under these conditions, the oscillator is inactive. The discharge of capacitor 342 that occurs when tube 333 becomes conductive results in a lowering of the grid potential of tube 336 sufficient to render tube 336 non-conductive. With tube 336 non-conducting, a 200 p/s sinusoidal output voltage appears on the grid in tube 337, and this voltage is applied across the resistor 343 on the control grid for tube 344 which is included in the oscillator square amplifier 97. The oscillator 96 has its resonant circuit adjusted so that it works with the same frequency as the oscillator arranged in the manager's office, i.e. of 200 p/s in a particular embodiment of this invention. The characteristic of this oscillator 96 is such that the sinusoidal output voltage is always preceded by a negative half period as shown by line I in fig. 8A.

Tube 344 of the square wave amplifier 97 is easily driven between conducting and non-conducting states by the sinusoidal voltage it] receives from oscillator 96. As a result, a substantially square wave is obtained as the output voltage from the anode of tube 344 when this tube is driven to blocking in the negative half period of its input voltage wave and to saturation at the positive half period of its input voltage wave. The voltage which thus appears on the anode of this tube is applied via capacitor 345 to the control grid in tube 346 which is included in the displacement pulse E—J flip-flop circuit 98, approximately as shown by line J in fig. A8.

Eccles-Jordan flip-flop circuit 98 is of the usual type comprising the two interconnected triodes 346 and 347. This trigger circuit has two stable states, one in which tube 346 is conductive with tube 347 blocked and another in which these relative conduction states are interchanged. At each positive voltage variation on the anode for tube 344, a positive pulse appears on the grid for tube 346, which pulse seeks to make this tube conductive with the result that tube 347 is blocked. Each negative voltage variation on the anode of tube 344 produces a negative trigger pulse on the grid of tube 346, causing this tube to be blocked and tube 347 to conduct. In this way, the flip-flop trigger circuit 98 is caused to operate alternately between its opposite states at a rate determined by the oscillator 96.

The output voltage appearing on the anode of tube 347 in the flip-flop circuit 98 is applied across wire 99 of the intermediate shift trigger circuit 100. As the output voltage of the oscillator square amplifier 97 rises, tube 346 becomes conductive and tube 347 non-conductive so that the voltage obtained from the anode of tube 347 also rises. The voltage applied to the grid in pipe 350 and across wire 99 on the grid in pipe 348 can be considered to be in phase with the voltage on the anode of pipe 344. This last voltage waveform is as shown at line J in fig. 8A. The wire 99 is connected directly to the control grid in the triode amplifier tube 348 which acts as an inverting amplifier so that the voltage on the anode of this tube is reversed in relation to that which was impressed on its control grid. The waveform of the voltage which appears on the anode of this tube 348 and is applied to the rail 116, corresponds to that shown by line L in fig. 8A.

The voltage on the anode of the tube 347 is also applied across a switching resistor 349 to the control grid for the triode amplifier tube 350 which is included in the pulse inverter 101. The pulse inverter 101 obtains at the connection for its anode load resistors 351 and 352 a voltage which is the reverse of that applied to its control grid. The reverse voltage is then applied directly to the control grid of the triode amplifier tube 353 which is included in the main offset trigger amplifier 102. This amplifier stage 102 also inverts its input waveform to provide an output voltage on rail 115 which is shown at line K.

The maximum voltage on both rails 115 and 116 must be limited to some previously selected values. If this is not done, the cathode of the tubes included in the shift register may be raised so high in relation to their anodes that they will be fired in the reverse direction. The maximum voltage that appears on rail 115 when pipe 353 is non-conductive is determined by the voltage at the junction between voltage divider resistors 354 and 355. Any transient voltage variation that seeks to appear on this rail and raise the voltage across it before selected maximum value induces a current across the rectifier 356 which is shunted to the anode load resistor 357. This prevents the voltage on rail 115 from rising above the previously selected value.

As shown in fig. 8A, the output of the oscillator square amplifier 97 is at a maximum during the negative half-cycles of the output of the oscillator 96 and is at a minimum during the positive half-cycles of the oscillator output. Each positive pulse edge shown in line J acts on flip-flop trigger circuit 98 to make tube 346 conductive. The first negative half period of the output from the oscillator 96 is thus certain to affect the flip-flop circuit 98 to the state where the tube 347 is non-conductive so that a high voltage is applied to the control grid of the tube 350. The conduction of the tube 350 which follows from this causes a low voltage appears on the control grid for pipe 353. The weakly conductive state of pipe 353 then causes this pipe to deliver a relatively high voltage to rail 115, as shown by line K in fig. 8A. At the same time, the high anode voltage for tube 347 causes tube 348 to become highly conductive so that a relatively low voltage appears on the intermediate displacement trigger rail 116, as shown by line L in fig. 8A. In each positive half period of the output of the oscillator 96, the relative conduction states of the various tubes are reversed so that the voltage on the main displacement trigger rail 115 is at a low level and the voltage appearing on the intermediate displacement trigger rail 116 is at a high level. The voltage on rails 115 and 116 consequently varies as the oscillator continues to operate, alternatively between two levels in the manner shown by lines K and L in fig. 8A.

Most system functions that are performed by the station step device 86 throughout a duty cycle relate to the influence of the displacement register. Before we discuss in detail how these different system functions are carried out, it is therefore considered expedient to describe the circuit arrangement and the working method of this shift register.

The shift register of the present invention is shown in detail in fig. 7C and comprises several steps, one of which is arranged for each of the fifteen pulse periods included in a station period with an additional step No. 0. In fig. 7C, only steps No. 0 to 3 and steps No. 14 and 15 are shown. The other steps in the shift register are identical to those shown.

Each shift register stage comprises two grid-controlled cold-cathode glow discharge tubes. The storage tube "REG" for each stage, such as tube 368 for stage no. 1, provides for the storage of the various code characters. The intermediate pipe "INT" for each stage, such as pipe 361 in stage #0, provides an intermediate or temporary storage of each character as it is transferred from one stage to the next.

Each tube in the shift register is controllable to one of two distinct states, i.e. a stable conductive state "1" for storing a marker code character and a stable non-conductive state "2" for storing a space character.

The cathode of each of the storage tubes "REG" is directly connected to rail 115, and the cathode of each intermediate tube is similarly connected to rail 116. The anode of each storage tube and each intermediate tube is connected across an anode load resistor with (B+). The grid for each tube is connected across a resistor and rectifier in parallel with a rail 362. The grid for tube 360 is e.g. connected across the parallel connection of the resistor 363 and the rectifier 364 with rail 362 which is connected to the connection point between the resistors 365 and 366. This voltage divider consisting of the resistors 365 and 366 is connected between (B+) and ground and provides a positive bias voltage for the grid for each tube in the displacement register for reasons that will be more fully described later. Resistor 366 in this voltage divider is shunted by capacitor 367 whose task is to prevent any sudden variation in the voltage on rail 362 which may be produced by transient variations in the (B+) power source.

Each of the storage tubes except the one included in stage No. 0 receives an input from the anode of the intermediate tube associated with the immediately preceding stage. For example, the storage tube 368 for stage no. 1 receives an input across capacitor 369 from the anode of the intermediate tube 361 which is included in stage no. 0. Each intermediate tube, on the other hand, receives an input on its grid from the anode of the storage tube in the same step. The grid of the intermediate tube 361 receives its input across capacitor 370 from the anode of the storage tube 360, i.e. in stage No. 0. In contrast, the storage tube of stage No. 0 receives an input from its grid across capacitor 371 from the gate-controlled shift register control marker amplifier 92 As will later be described in detail, a positive trigger pulse is obtained from this gate-controlled amplifier 106 for each marker character that appears in the control code for that particular field station.

The displacement rails 115 and 116 are alternately pulsed with positive voltages at a rate determined by the oscillator 96. Normally, however, as shown in FIG. 8A at lines K and L, rail 115 is at a low potential and rail 116 is at a high potential. This means that the cathode of each storage tube is at a low positive potential relative to ground. A comparison of lines E, G and L in fig. 8A shows that the delayed selection amplifier 79 of FIG. 7A provides a positive output pulse for each mark in the control code, and that these trigger pulses are suitably delayed so that each occurs at a time when the voltage on the main displacement rail 115 is at a low level.

The first of these trigger pulses occurring during a station period momentarily raises the voltage on the grid in tube 360 sufficiently relative to the grid to cause this tube to light. In cold-cathode tubes of this type, the grid must be operated positively in relation to the cathode of the tube to be ignited. As each of these tubes is provided with a positive bias from rail 362, one only needs a positive pulse of relatively moderate amplitude on the control grid to cause the tube to ignite. The positive bias must of course not be so great that it causes the tube to light up, when no external pulse occurs.

The conducting state for pipe 360 only exists temporarily as the voltage on rail 15 soon rises to a sufficient value in relation to the anode for this pipe, so that the line can no longer be maintained. As a result, the tube extinguishes and the voltage at its anode which was at a low level while the tube was conducting, suddenly rises to a higher level approximately equal to the (B+) voltage.

At the same time as the voltage on rail 115 is raised, the voltage on rail 116 is lowered. Consequently, the positive voltage variation on the anode of tube 360 which is applied across the capacitor 370 and is effective in producing a positive pulse on the grid in tube 361, will raise the grid voltage on tube 361 sufficiently in relation to the now approximately earthed cathode for the tube 361 to ignite.

A short time later, the voltage on rail 116 is raised at the same time as the voltage on rail 115 is lowered. The resulting low anode voltage on tube 361 is not sufficient to keep this tube conducting, so that it is extinguished. Its anode voltage then suddenly rises so that a positive trigger pulse appears on the grid of tube 368 which is the storage tube of stage #1. This pulse can now cause tube 368 to become conductive, as the cathode of this tube is at a low potential due to the low potential now appearing on rail 115. Any tube in the shift register that is not in the conducting state is not affected by the rise in voltage of the associated shift rail, so that it cannot provide a pulse to fire another such tube .

The speed at which the displacement pulses are applied to the rails 115 and 116 is determined by the oscillator 96 in fig. 7B, as already mentioned, and this oscillator is designed to operate at the same frequency as oscillator 35 in the executive office. In this way, the occurrence of shift pulses is synchronized with the received control code symbols in such a way that rail 115 is always at a low potential level, as each pulse is applied to the tube 360 of the first shift register stage from the shift register's control marker gate-controlled amplifier 92.

Between successive pulse periods, the displacement pulses first cause the potential of rail 115 to rise, thereby causing any storage tube in a conductive state to transfer this state to its associated intermediate tube, and

then it will rise in potential on rail

116 as follows, cause the conductive state of any intermediate tube to be transferred to the storage tube in the following step.

From this description it will be seen that the two displacement pulses occurring between successive pulse periods cause the particular state of each displacement stage to be transferred to the next higher numbered stage in the sense that the conducting state of a storage tube results in the conduction of the storage tube in the second-following stage, while any storage tube in a non-conducting state transfers no input to the next stage, so that the next stage remains non-conducting.

As an example of this working method, it is shown by line N in fig. 8A that the first positive trigger pulse obtained from the gate controlled amplifier 92 and applied to the first stage of the shift register occurs at a time when the voltage on rail 115 is at a low level (see line J) and causes conduction of the storage tube in the first step, as shown at line 0 in fig. 8B. Shortly thereafter, the potential for rail 115 is raised so that the storage tube in stage no. 0 is extinguished, and this results in the ignition of the intermediate tube in this stage, as indicated by line P. Correspondingly, the potential on rail 116 is raised, while the voltage on rail 115 is lowered to its original level. This results in the extinguishing of the intermediate tube in the first stage and the ignition of the storage tube in the second stage. At this time, no pulse is applied to the storage tube for the first stage from the gate-controlled amplifier 92, so that this tube is not ignited. When the voltage on rail 115 is raised again, no output is pressed from the storage tube to ignite the associated intermediate tube for the first stage.

A single pair of displacement pulses appears on rails 115 and 116 at the end of

the last pulse period, because the generation of these pulses can only be terminated by the occurrence of the next station call pulse.

If the shift register only included

one step for each pulse period, the first received character would be stepped out of the No. 15 step at the last pair of shift pulses.

By providing the additional No. 0 stage, the first received character is stepped into stage No. 15 at the last pair of shift pulses and the last received character is then transferred from stage No. 0 to stage No. 1.

It can therefore be considered that the application of shift pulses to all stages of the shift register in successive pulse periods, while a time-shifted code of input pulses selectively ignites the storage tube in the first stage, will cause the time-shifted code to be entered into the shift register. However, at the completion of the scrambling operation, the first received character from the gated amplifier 92 remains in the last stage of the shift register, while the last received code character is in the first stage.

This stepping operation which is produced by the application of displacement pulses on the rails 15 and 116 can, as in the present embodiment of this invention, also be used for stepping an indication code out of the displacement register. At the beginning of a drive period, a positive trigger pulse is thus obtained from a gate-controlled amplifier and applied across contacts on one of the groups of output relays to the various shift register stages no. 1 to 15. At the start of the drive period in which group A controls are received, e.g. . a positive trigger pulse from the shift register indication marker's gate-controlled amplifier (group A) 106 and is applied across rail 107 and across contacts on group A indication relays 108 to the respective shift register stages. The manner in which this pulse is provided in response to a control received from the station stepping device 86 will be described later in detail.

These different relays can be selectively controlled devices schematically shown at block 21 in fig. 6B. For each indicating relay that is in an affected state, the positive trigger pulse appearing on rail 107 is applied across a working contact for such a relay and a series-connected neon lamp on the grid in the storage tube for a corresponding step of the shift register. If e.g. the relays 375 and 377 in group A with indicating relays are affected, the positive trigger pulse is applied to rail 107 over working contacts 380 and 381 respectively for these relays as well as over the respective some-lamps 382 and 383 to the storage tubes 368 and 385 included in the shift register steps no. 1 and 3. Similarly, if group A of indicating relays 376, 378 and 379 drops, their working contacts will be opened so that the positive trigger pulse on rail 107 cannot be applied to make the storage tubes of the second, fourteenth and fifteenth shift register stages conductive. In a following station period, when Group B controls are received at the field station, similar positive trigger pulses on rail 127 are then actuated to selectively fire the storage tubes in the various shift register stages in accordance with the actuated state of the various Group B indicating relays.

With the storage tubes of the various shift register stages selectively actuated in this manner in accordance with an indication code at the beginning of a drive period, shift pulses on rails 115 and 116 during that drive period act to step indication code characters from stage to stage. The result is that the final stage No. 15 has its tubes selectively made conductive and non-conductive in successive periods in a time-shifted pattern exactly corresponding to the special indication code placed in the shift register at the beginning of the station period. This transfer of the code characters from one stage in the shift register to the next is carried out in the same way as the transfer of control characters which are printed in a time-shifted code pattern to the first stage. This activity has been described in detail so that it is not considered necessary to describe it again in detail in connection with the indication code.

It will now be described how different system functions relating to the influence of the displacement register are carried out by the station step device 86.

In the special station period in which group A controls are received from the management office and group A indications are sent back to the management office, the annealing is used on cathode Kl for the multi-cathode tube in the station step device. The positive voltage that appears on this cathode Kl is applied across a separation rectifier 390 and across the capacitor 330 on the wire 331. From the wire 331, this positive voltage is applied to the input of the oscillator gate control 95 in the same way as already described in detail. This causes the oscillator 96 to become active with the end result that displacement pulses appear on the rails 115 and 116.

The positive voltage pulse that appears on the wire 331 is also impressed on the control grid for the gas discharge tube 391 which is contained in the shift register control amplifier 92. As the capacitor 330 has a relatively large value, the time constant for charging this capacitor is correspondingly long, so that the positive voltage obtained from the cathode Kl causes the grid potential of tube 391 to remain higher than the normal negative bias potential provided by the grid across resistor 392 being connected to (B—) for almost the entire station period. It can therefore be seen that throughout the station period during which the annealing is on cathode Kl, the control grid on the gas discharge tube 391 is positively gate-opened, so that it can become conductive in response to a positive pulse that is applied to its screen grid.

The screen grid for tube 391 receives the positive trigger pulses shown at line D in FIG. 8A. These are obtained from the delayed selection amplifier 79 in fig. 7A and the screen grid is pressed over the capacitor 393. The screen grid is normally negatively biased by being connected via resistance 394 to the (B—) voltage source. Each positive trigger pulse applied to the screen grid raises its potential sufficiently so that the tube will be lit provided the control grid is correctly gate-opened during this station period.

The anode for tube 391 is connected across a resistor 395 and parallel capacitor and primary winding for transformer 396 with (B+). The inductive effect of the voltage in the primary winding on transformer 396 in the anode circuit of tube 391 results in the generation of an induced voltage in the anode circuit of this tube whenever it becomes conductive, and this induced voltage seeks to momentarily reduce the anode voltage to such a low value that the line is no longer maintained and the tube is then extinguished. As a result of this self-extinguishing characteristic, gas tube 391 is only momentarily conductive for each trigger pulse applied to its screen grid.

Whenever the anode of tube 391 conducts current, a voltage is induced in the secondary winding of transformer 396. A positive voltage variation is then impressed across wire 397 on the grid circuit of the storage tube 360 arranged for shift register stage No. 0. Any negative voltage variation induced in the secondary winding on transformer 396 and seeks to act on wire 397, is effectively short-circuited by rectifier 398 which is connected directly across the terminals of the secondary winding on the transformer. The wire 397 is connected across the brake diode rectifier 403 to the connection point for the voltage divider resistors 399 and 400 which are connected in series between (B+) and ground. This causes the upper level of the voltage on wire 397 to be slowed down to the level of the voltage at the connection point of these two resistors 399 and 400, thereby preventing the voltage on wire 397 from exceeding a suitable value. Capacitor 401 shunting resistor 400 seeks to prevent any sudden change from occurring on wire 402.

During the time when cathode Kl in the station step device 86 is at a high potential, a positive gate voltage is applied across wire 405 and across capacitor 406 to the control grid for the gas discharge tube 407 which is included in the shift register indication's marker gate-controlled amplifier (group A) 106. This control grid becomes also normally negatively biased so much as to prevent the tube from firing at a junction arranged across resistor 408 to (B—). The positive voltage from cathode Kl raises the control grid potential for tube 407 throughout the station period in which the glow is on cathode Kl, so that the tube can be ignited provided that the potential on its normally negatively biased screen grid is raised. The screen grid of this tube 407 is connected across capacitor 409 and across wire 410 to the anode of tube 287 which is included in the sync and station call pulse multivibrator 81. Normally the anode of tube 287 is at a high voltage level because this tube is normally in a non-conducting state . However, upon the occurrence of each sync or station call pulse in the cycle, tube 287 becomes conductive for a short interval so that its anode voltage is reduced to a low level. Capacitor 409 is then discharged and a negative voltage pulse appears on the screen grid of tube 407, but this negative pulse of course has no effect on the conduction in tube 407. When the multivibrator 81 returns itself to its normal state so that tube 287 once again becomes non-conductive , its anode voltage rises suddenly so that capacitor 409 is charged with the result that a positive pulse appears on the screen grid for both tubes 407 and 411. Of these two tubes, only tube 407 has its control grid positively gate-biased in that the glow is on cathode Kl in the station step device so that only tube 407 is lit at this time. Tube 407 remains lit only momentarily as it is arranged to extinguish itself in the same manner as described for tube 391.

In a similar manner as described for the gate controlled amplifier 92, the ignition of tube 407 causes a positive pulse to be applied across wire 107 to the contacts of the group A indicating relays shown in FIG. 7D. As described in connection with the influence of the shift register, this pulse on wire 107 causes various stages of the shift register to be properly prepared at the beginning of the station period in accordance with the desired code pattern to be sent to the head office during this station period. The connection in relation to the secondary winding of transformer 404 is somewhat different from what it was for transformer 396. Transformer 404's lower terminal on the secondary winding is connected directly to the connection point between resistors 399 and 400. A positive bias therefore occurs on rail 107 and this bias helps with the ignition of the various cold cathode tubes in the displacement register.

In the two station periods when the annealing is on the cathodes K10 and K9 in the station step device 86, a positive voltage is applied across the capacitor 412 to the grid for the triode 413 which is part of the group A control design amplifier 120. The control grid in tube 413 is normally biased to blocking in that it across resistor 414 is connected to (B—). During the two station periods during which the annealing is on cathodes K10 and K9, however, the grid potential for tube 413 is raised sufficiently to cause this tube to become conductive. As a result, current flows from (B+) across 415 and also from (B+) across the winding of relay 121 across resistor 416 and the anode-cathode circuit for tube 413 to ground. As a result, this relay is energized at the beginning of the station period in which the glow is transferred to cathode K10 and remains energized throughout this station period and the following one until the glow is transferred to cathode K8. The effect of the attraction of this relay 121 when executing the controls stored in the shift register will later be described in detail.

No system function is provided when the glow of the station step device 86 is on cathode K8. At some of the field stations, however, a connection is made from this cathode K8 to the station stepping device's synchronization control 83 for reasons already described.

In the following station period, when the annealing has been transferred to cathode K7, a positive gate voltage is applied across wire 325 and across capacitor 417 to the control grid in the gas discharge tube 418 which is included in the clearing pulse amplifier 122. The grid in this tube 418 is connected across a resistor 419 with (B— ). The cathode and screen grid are both connected to the connection point of resistors 420 and 421 which are connected in series between (B—) and ground and as a result of this connection both cathode and screen grid are held at a potential substantially negative with respect to ground. Capacitor 422 shunting resistor 421 seeks to prevent any sudden voltage variation that may be produced by transient variations in the voltage source (B—) from acting on the cathode and screen grid of tube 418.

Although the cathode of tube 418 is held at a negative potential, the control grid, being directly connected across resistance, is. 419 with (B—), still more negative so that the tube is normally biased to blocking. However, when the potential on cathode K7 is raised, the voltage of the control grid on tube 418 is raised sufficiently in relation to the cathode voltage, so that this tube is lit. Due to the relatively large negative voltage that appears on the cathode, the anode voltage also becomes substantially negative when the tube is ignited, so that the grid voltage for tube 353 which is included in the main displacement trigger amplifier 102 is suddenly driven to blocking. Normally, this tube 353 is conductive so that a small voltage appears on rail 115 which leads to all storage tubes in the displacement register. When tube 353 is blocked by the ignition of tube 418, the anode voltage for tube 353 is raised to a high value so that the cathode voltage for each of the storage tubes in the shift register is raised to such an extent relative to the anode that the line cannot

is maintained* and they all extinguish. A high

voltage appears on rail 116 and this

time also so that none of the intermediate tubes in the shift register, whose cathodes are connected to rail 116, can become conductive when an associated storage tube is switched off. In this way, the blocking of tube 353, so that rails 115 and 116 are both at a high potential, ensures that all storage and intermediate tubes in the shift register are returned to non-conductive states. Any previously received control code is thus canceled from the shift register, so that the shift register is properly prepared to receive a new indication code.

The system functions that are performed when the annealing is on cathode K6 in the station step device 86 are the same as those that occur when the annealing is on cathode Kl. Positive gate voltages are thus applied across capacitor 330 and across wire 331 to the oscillator gate control 95 and to the shift register's control marking gate-controlled amplifier 92. A positive gate voltage is also applied across wire 425 to the shift register indication marker gate-controlled amplifier (group B) 126. Thus, when the screen gate of tube 411 is supplied with a positive gate voltage by a trigger pulse received from the sync and station call pulse multivibrator 81 at the beginning of the station period , a pulse is produced on wire 127, and this pulse is then selectively applied across contacts for this group's B indicating relays in fig. 7D to the various stages of the shift register. In this way, an indication code is printed on the shift register whose characters correspond to the influencing states of the group B indication relays.

During the two following station periods, when the annealing is on the cathodes K5 and K4, for the station step device 86, a positive gate voltage is applied across wire 426 and across capacitor 427 to the control grid of tube 428 which is included in the group B control amplifier 130. This gate voltage makes tube 428 conductive so that an anode current flows in this tube and through the winding on the group B control relay 131 in fig. 7C. In these two station periods, relay 131 is thus attracted so that group B controls which are stored in the shift register can be realized in a way that will now be described.

No special system function is carried out during the station period during

which the annealing is on cathode K3. At certain field stations, however, as shown in fig. 7A, a connection is arranged from this cathode K3 to the station stepping device's synchronization control 83.

During the next drive period, when the glow is on cathode K2 of the drive step device 86, a positive gate voltage is again applied to wire 325 to tube 418 of the clear pulse amplifier 122. Exactly as in the manner already described, the ignition of tube 418 causes the voltage on rail 115 is raised to a high value at the same time that the voltage on rail 116 is already at such a high value that all tubes in the shift register are brought back to their normal non-conductive states and the shift register can be considered cleared for any previously received control.

An overview of the various system functions is given in the following form:

Each control code received at a field station and entered into its shift register is subsequently executed by selectively controlling various control relays in accordance with the states of the various shift register stages. After Group A controls are received, the Group I A control relays are made receptive to the conditions of the shift register stages. The group B control relays are similarly made receptive to the conditions for the shift register stages immediately after the group B control codes have been entered into the shift register.

As described in connection with the control matrix and control arms of the executive office apparatus, the special code characters sent in the first two pulse periods of a station period can be determined in accordance with the position of a two-position switch or control arm. This coupler selects whether a marker should appear in the first or second pulse period in the code. A space character is then transmitted in the remaining pulse period. These first two characters in the control code are optionally stored in the shift register steps no. 15 and 14 in accordance with the description already given for the shift register. It will be shown how the codes for these last two stages of the shift register can be made effective to control the operations of a single control relay in either the group A or group B set of control relays.

A three-position control arm can also be easily used as described to select in which of the three successive pulse periods a marking character is to be sent. The detailed circuits in the management office devices show how such a three-position control arm determines the structure of the control code sent in the last three pulse periods of a station period. As the last three code characters are possibly stored in the shift register's stages no. 1, 2 and 3, it is shown in fig. 7C how these three stages of the shift register can be used to control the states of two of the control relays in a set of control relays.

Each of the relays controlled by the shift register is shown in fig. 7C as a magnetic rod type relay. The energization of the upper winding causes the relay to be attracted, while the energization of the lower winding causes it to drop. The magnetic rod characteristic causes each relay to remain in its last affected position until it is controlled to an opposite position.

Each control tube in a shift register stage has its anode connected across a resistor of (B+>. The anode of tube 360 of shift register stage No. 0 has its anode connected in this way across resistor 435 of (B+). The anode of tube 368 is also connected across a register 436, across the upper winding of relay 2RGZ and rest contact 430 of relay 2LGZ with rail 437. Yet another connection is arranged from the anode in pipe 368, across resistor 438, the upper winding of relay 1RGZ and rest contact 431 of relay VLGZ to rail 439. Similarly, the upper windings for the relays 1LGZ and 2LGZ are connected across respective resistors with the anode in tube 385 which is included in the shift register stage No. 3.

The anode in tube 384, which is the storage tube for shift register stage No. 2, is pre-

tied across resistance 440 and across lower windings of relays 2LGZ and 2RGZ in parallel with rail 437. This anode for tube 384 is similarly connected across resistance'441 and the lower windings in parallel of relays 1LGZ and 1RGZ with rail 439.

If it is assumed that the three-position control arm at the manager's office is set so that marker characters are sent in the fifteenth pulse period so that the space characters are sent in the fourteenth and thirteenth pulse periods, then the shift register in the corresponding field station will have its stages No. 1 to 3 prepared so that tubes 368 will become conductive while tubes 384 and 385 will become non-conductive. If it is further assumed that it is group A controls that have been received, then it will be

clearly consistent with the description already given with regard to them

different system functions which are arranged by the station step device 86, that in the two station periods immediately after the reception of such group A controls, the group A control execution relay 121 will be attracted. When relay 121 is attracted, it closes its working contact 442 to energize rail 439. A current then flows from (B+) across working contact 442, rail 439, rest contact 431 of relay 1LG°, the upper winding of relay 1RGZ, resistor 438 and over the anode-cathode circuit of the now conducting tube 368. This tube has its cathode connected across rail 115 and this rail is connected to the anode of tube 353 so that a path is provided from the anode of this tube 353 to earth to complete a circuit which includes the upper winding of relay 1RGZ. As a result of this energisation, relay 1RGZ is attracted.

If the Group A control code were arranged so that the marker was received in the thirteenth instead of the fifteenth pulse period, tube 385 in step #3 would become conductive, while tubes 368 and 386 would then become non-conductive. Under these conditions, a circuit would be closed by the energization of relay 121, and this circuit would cause current to pass through the upper winding of relay 1LGZ to energize this relay.

If the control code were arranged so that a mark was sent in the fourteenth pulse period and intermittently occurring in the thirteenth and fifteenth pulse periods, so that tube 384 would become conductive, then a circuit would be completed by energizing relay 121 to energize the lower windings of both relays 1RGZ and 1LGZ. The energization of these two lower windings would result in both of these relays dropping. In this way, three successive characters of the control code can be used for the control of two relays in a field station under any of three special conditions. If we follow the reception of the group B controls in any field station, group B control execution relay 131 is attracted. Energy is then applied to rail 437 so that relays 2RGZ and 2LGZ are prepared for the states of the first three shift register stages in exactly the same manner as just described for relays 1RGZ and 1LGZ.

Any two shift register stages, such as Nos. 14 and 15, may be used to control a single relay in any of the Group A or B sets of control relays. If a mark is sent in the first pulse period, so that storage tube 445 in step no. 15 becomes conductive, then the lower winding in relay 1NWR or 2NWR is energized depending on whether it is a group A or a group B control that has been done. Similarly, the upper windings of these relays can be selectively energized if a mark is sent in the second pulse period to cause tube 446 in stage #14 to become conductive.

The carrier wave transmitter 24, shown in fig. 7D, is modulated directly from the last stage of the shift register and becomes effective so as to cause the special carrier frequency F2 to be selectively imprinted on the line wires in a special coding pattern with markings and spaces.

The carrier wave transmitter 24 is in most respects the same as the carrier wave transmitter which is arranged in the manager's office and shown in fig. 4C. The carrier wave transmitter 24 also comprises a pulse-driven Hartley oscillator comprising tubes 443 and 444. Tube 443 is normally conductive because its control grid is connected across resistor 447 with (B+). With tube 443 conducting, the oscillator is inactive, so that the carrier frequency F2 cannot be applied to the line wires 10.

The control grid in tube 443 is connected via a coupling capacitor 448 with the anode for cold cathode intermediate tube 458 in displacement register stage no. 15, shown in fig. 7C. Normally, tube 458 is in a non-conducting state, so that its anode voltage is at a relatively high level, and capacitor 448 is then charged to this high voltage that occurs between the anode of tube 458 and the grid of tube 443, which is approximately at ground potential.

The last stage of the shift register is selectively driven to its "0" or its "1" state in successive pulse periods in accordance with the indication code previously stored in the shift register which is stepped out, while a control code is stepped in. For each space character in the indication code, tube 458 remains non-conductive, because then there is no change in the state of tube 445.

In each pulse period a marker appears in the last stage of the shift register, tube 445 is lit. Then the rise in potential across rail 115 causes tube 445 to be extinguished, and tube 458 is then lit. Each time this tube 458 is ignited in this way by the occurrence of a marker character in the indication code, a lowering of the anode voltage of this tube occurs throughout the entire pulse period. As a result of this, there is a sudden drop in voltage on the grid in tube 443 and this voltage drop is sufficiently large and of long enough duration to keep tube 443 non-conductive throughout this pulse period. As a result, the oscillator becomes active, causing the anode current in tube 444 to vary sinusoidally at a frequency determined by the flywheel circuit in the cathode circuit for tube 443.

Tube 444 is provided with a load resistor in both its anode and its cathode circuit, so that the required out-of-phase voltages can be obtained by driving the reverse cycle power amplifier comprising tubes 449 and 450. The grid in tube 449 receives e.g. an input voltage produced across anode load resistor 451 in tube 444, while the input grid voltage in tube 450 is obtained from the voltage drop across series resistors 452 and 453 in the cathode circuit for tube 444.

The grids in the pipes 449 and 450 are also connected across the respective resistors 454 and 455 with a tap on the potentiometer 456 which is connected between (B-) and ground. The tap on the potentiometer is set to provide the desired bias for tubes 449 and 450 and thereby provide a means of regulating the output level of this amplifier stage. The primary winding for a transformer 457 is included in the anode circuits for tubes 449 and 450 and a tap on this primary winding is connected directly to (B+). The alternating current which is induced in the secondary winding and has a frequency F2, is applied across the bandpass filter 25 to the line wires 10.

As only half of the pulse period depends on the transmission of the code character to

tube 458 before the potential on rail 116 is raised to extinguish this tube 458, the tube remains

in its leading state only in one half of

a pulse period. Since the carrier transmitter 24 is operative to deliver the carrier frequency F2 to the line wires only during the time that tube 458 is conductive, it follows that indicating

the code comprises the marking characters which are the same as those appearing in the control code, in that they comprise a pulse period in which the carrier frequency is transmitted in the first half but without any carrier frequency transmitted in the second half.

The indication code comprising intermediate pulses of the carrier frequency F2 is printed from the line wires 10 above the bandpass filter 26 on the control grid for the amplifier tube which is included in the carrier frequency receiver 27 in the manager's office (Fig. 4D). Both tube 464 and tube 465 are actuated as cathode-biased class A amplifier stages with resistive capacitance coupling applied between both stages. The anode voltage for tube 465, which comprises intermediate pulses with carrier frequency F2, is applied across a coupling capacitor 466 to the cathode of a diode rectifier 467. The filtering of the rectified output that occurs on the anode of this rectifier 467 is ensured by a filter comprising resistors 468 and 469 as well as capacitors 470 and 471. As a result of this rectification and filtering, a voltage is applied to the control grid of triode amplifier tube 472 across resistor 473 which has a negative pulse for each occurrence of a carrier frequency pulse applied to the receiver input.

The square wave amplifier 140 is essentially an overdriven amplifier whose control grid is connected across a resistor 473 with the connection point of the resistors 474 and 475, which are connected in series between (B+) and ground. Because of the positive bias applied to the control grid in this way, tube 472 is normally in a fully conducting state, so that there is a high voltage across its anode load resistor 476. For each occurrence of carrier frequency F2 in the received indication code, the control grid is driven negative so that the tube becomes blocked, and its anode voltage rises to a high value. Since only a relatively small voltage variation on the control grid of this tube 472 causes the tube to vary from its normal conducting state to a blocking state, the variations in anode voltage in tube 472 are relatively sudden with the result that an essentially square voltage wave appears on the anode for this tube and is pressed over the coupling capacitor 477 on the control grid for tube 478 which is included in the marking multivibrator 141.

Tubes 478 and 479 are included in a "one-shot" multivibrator which is normally in the state where tube 479 is conducting and tube 478 is non-conducting. In response to the positive leading edge of each pulse appearing on the anode of tube 472, the conducting state of these two tubes 478 and 479 is suddenly reversed. This condition is shown in fig. 5B, which shows at line V that the marker multivibrator 141 reverses its state at the leading edge of each pulse delivered by the receiver 27. This negative leading edge of the receiver pulse coincides with the positive leading edge of the output pulse delivered by the square wave amplifier 140. After a short interval which is less than the 2-y2 millisecond interval that the carrier frequency F2 is applied for an indication mark, the mark multivibrator 141 returns itself to its normal state. The resulting negative trailing edge of the voltage appearing on the anode of tube 479 of this multivibrator produces a negative trigger pulse on the grid of tube 480 of trigger delay 142.

The trigger delay comprises a differentiator circuit similar to that already described. Tube 480 becomes non-conductive for a time in response to the negative pulse impressed on its grid across capacitor 481, and during this interval of non-conduction, the anode voltage in tube 480 rises to approximately the (B+) level, because then there is no some anode current drop across its load resistance 482. As shown in fig. 5B, line W, as a result, a positive trigger pulse is provided by the trigger delay to rail 143 each time the marker multivibrator 141 returns itself to its normal state in which tube 478 is non-conducting.

The indication storage matrix 30 shown in fig. 4E comprises a number of sets of cold cathode storage tubes. One such set of cold cathode storage tubes is arranged for each station period of a duty cycle and is used to store all the indications received in each cycle during that particular station period. In reality, therefore, two such sets of storage tubes are arranged for each field station in the system. One batch is arranged to store group A indication codes from this station and the other batch is arranged to store group B indication codes from this station.

Each such batch of storage tubes comprises an individual tube for each of the fifteen pulse periods included therein

station period. The cathode of each storage tube is grounded and its anode is

connected across an anode resistor and the primary winding of a saturable transformer with a rail. The anode for this tube 494 is thus connected across a resistor

498 and the primary winding for the transformer 499 with rail 500. Rail 500 is normally at a relatively high potential at

that it is connected to the cathode of the cathode follower amplifier tube 506 which is included in the indication clarification amplifiers 144.

Each of the storage tubes is gate controlled from three different sources, and all three gate voltages must be simultaneously present before any tube can be fired. In any station period, the correct one is selected from among all the sets of storage tubes by a gate voltage obtained from the indication matrix gate amplifiers 146.

Each of these gate amplifiers comprises a triode amplifier tube which has its cathode connected to ground and its grid connected to a corresponding one of the steps in the station stepping device 44 of fig. 4B. The special gate amplifier tube 487 which pertains to group A station no. 1 station period has e.g. its control grid connected across resistor 485 to the right anode of the Group A Station No. 1 station stepper stage 44. During the Group A Station No. 1 station period, this particular stage of the station stepper device is in its "1" state, so that the the right half of this tube 486 is conductive. Consequently, a relatively low anode voltage will appear on the right anode in tube 486, and this voltage is sufficiently low that it cannot overcome the negative bias that is provided for the control grid in tube 487 by being connected across resistance 488 with (B— ). Tube 487 is therefore blocked so that its anode voltage is at a high level. This high voltage is then applied across rail 489 to each of the tubes that are included in group A station no. 1 set with storage tubes, and provides one of the required gate voltages for these tubes. A connection is provided from rail 489 across a dividing rectifier to the grid for an associated cold-cathode storage tube. It is e.g. arranged a connection from rail 489 over separating rectifier 490 to the control grid for cold cathode tube 491 which is the storage tube arranged for controlling the cathode sign received in the first pulse period in this particular station period.

In each station period of a cycle, the group A station No. 1 stage for the station stepper is in its "0" state, so that the right half of tube 486 is non-conductive. The anode voltage, which is applied under these conditions to the control grid in tube 487 which is included in the gate amplifiers 146, raises the grid voltage for tube 487 sufficiently to make this tube conductive. Its anode voltage is then lowered so that a low voltage appears on rail 489, and this voltage is not of sufficient amplitude to properly open the grids for the various storage tubes connected to this rail. Similarly, the various gate amplifiers 146 are effective for

to choose the right batch of storage tubes

concerned, the special station period which is then effective.

The particular storage tube among each batch of storage tubes to be responsive to receive indication characters occurring in any pulse period is selected by providing additional gate voltages to the storage tubes from the cathode followers associated with the various stages of control stepper 43. In each pulse period during a drive period, one of the stages Nos. 1 to 15 of the control step device 43 is in its "1" state with the result that the corresponding cathode follower connected to that stage of the control step device supplies a high output voltage to one of the rails which supplies it to the indication storage matrix 30. Consequently, with step No. 1, the step for the control step device in its "1" state during the first pulse period of each station period, the rail 492 shown in FIG. 4E, its potential raised to a relatively high value. In the second pulse period in any station period, the potential of rail 492 is brought back to its normal value, and at the same time the voltage on rail 493 is raised to a higher level.

Thus, when any tube receives the required gate voltage, this tube can actually become conductive only provided it also receives a positive trigger pulse from rail 143. From the description that has been given, it is clear that such a positive trigger pulse will appear on rail 143 only if a marker has been received in the particular pulse period of the indication code. For example during the second pulse period for Group A station No. 1 station period, storage tube 494 simultaneously receives gate voltages from rails 493 and 489. If during this pulse period a positive trigger pulse also occurs on rail 143, the cumulative effect of all three of these voltages occurring on these rails raise the lattice potential of tube 494 to a sufficiently high value so that this tube will become conductive.

Fig. 5B shows how the different storage tubes in a batch contained in the matrix are selectively made conductive by a received indication code. A time-shifted pattern of indication marks is shown, each resulting in a positive trigger pulse from the trigger delay. These indication pulses are received during a group B station no. 5 station period, which is evident from the fact that the station step device then provides a high output voltage from its group B station no. 5 stage, as shown by line L in fig. 5A. Because of this, only the set of storage tubes that pertain to this particular station period receives the requested station period gate voltage.

In each pulse period for this station period, one of the pipes in this rate corresponding to the period receives a different gate voltage. At the time when the first output pulses shown as being delivered by the trigger delay (line W in Fig. 5B), deliver e.g. step no. 1 step for the control step device a high output voltage as indicated by line F in fig. 5A. In this way, only the step #1 tube in this particular batch receives the required second gate voltage so that only this tube can be fired by the output pulse from the trigger delay, which pulse is simultaneously applied to all the storage tubes for the matrix. Likewise, each of the following output pulses from the trigger delay results in only the corresponding storage tube of those included in the matrix being made conductive. The received indication pulses then result in selectively making the various tubes in the storage array conductive in the manner indicated by lines X to HH in FIG. 5B.

Each of the storage tubes, such as tube 494, is connected to one of the gate rails via a dividing rectifier. Pipe 494 is thus connected over rectifiers 495, 496 and 497 with rails 493, 143 and 489 respectively. These rectifiers do not prevent desired supply conditions from occurring. The rectifier 495 f. e.g. prevents a positive gate voltage on rail 143 from also appearing on rail 493 and thus being applied to all steps no. 2 in the storage tubes in the matrix.

When a storage tube in the matrix 30 is in its normal state, no anode current passes through the tube. When tube 494 is in its normal state, e.g. may no anode current flow through its anode resistor 498 and through the primary winding of the saturable transformer 499 and rail 500. Under these conditions, the reactance provided by the secondary winding of the saturable transformer 499 is high. Accordingly, the alternating current voltage applied to the circuit encounters a sufficiently high impedance so that there is insufficient current through the corresponding indicator lamp 502 to cause this lamp to light.

However, when the storage tube 494 has been made conductive by the appearance of an indication mark at the correct place in the indication code, an anode current flows through the primary winding of the saturation transformer 499 and this results in a substantial lowering of the reactance of the secondary winding. The impedance in the secondary circuit will consequently be lowered so that the voltage supplied by the alternating current source can supply a sufficient current to light the lamp 502. In this way, a visible indication of the indication storage matrix 30 is obtained by means of light in a lamp for each conductive storage tube . A somewhat similar circuit arrangement can be arranged so that each conductive storage tube can cause the energization of a corresponding relay.

At the beginning of each such station period, all previously stored indications of the particular set of storage pipes to be opened in order to react to the indications to be received in this station period are clarified. This clarification of indications is carried out by restoring the normal non-conductive state of each of the storage tubes in this particular batch and this is carried out by means of indication clarification amplifiers 144.

The indication clarification amplifiers 144 comprise two amplifier tubes for each station period and thus for each batch of storage tubes in the matrix 30. The clarification amplifier concerned with the storage batch for group A station no. 1 station period comprises the pentode tube 505 and the triode amplifier tube 506. Both the control and capture grids for tubes 505 are normally biased so much that they prevent the tube from becoming conductive. The bias voltage for the control grid is supplied by means of a connection across resistor 507 to rail 508. Normally rail 508 is at low voltage due to the wire provided from this rail 508 across resistor 509 to (B—). The trap grid for pipe 505 is connected across a resistor 510 with (B—) to achieve the desired blocking bias for this grid. The required operating voltage for the screen grid for this tube 505 is supplied by a connection across resistor 511 to (B+).

During the Group A station No. 1 station period of one cycle, the cathode follower connected to the corresponding step in the station step device 44 supplies a relatively high output voltage across the capacitor 512 to the trap grid in tube 505.

Rail 508 is supplied with a gate voltage from the station call rest step cathode follower of FIG. 4B across the capacitor 513 when this particular step for the control step device 43 is in its "1" state.

Since tube 505 can only become conductive when the potentials of both its trap grid and control grid are suitably elevated with respect to the grounded cathode, it is clear that this tube can become conductive only during the five millisecond pulse period of Group A Station No. 1 station period during which station call the rest step for the steering step device is in its "1" state.

When tube 505 is in its normal non-conductive state it delivers a relatively high anode voltage and cathode follower tube 506 is fully conductive so that its cathode is at a relatively high voltage, which is applied across rail 500 to the anode of each of the storage tubes in matrix 30 which is in group A station no. 1 the batch with storage pipes. As a result, these tubes are normally supplied with the required high anode voltage so that they can be selectively made conductive.

When the pentode tube 505 momentarily becomes conductive at the beginning of the Group A station No. 1 station period, the normally charged capacitor 515 discharges and reduces the grid voltage of tube 506 so that this tube blocks. The cathode voltage for this tube is lowered so much that there is no longer sufficient voltage on rail 500 to maintain the tubes in Group A Station No. 1 the set of storage tubes in a conductive state so any tube that has been controlled so is conductive in accordance with the received indication code in the previous cycle is returned to the non-conductive state. In this manner, the indication clearing amplifiers 144 cause each set of storage tubes to be cleared for all previously received indications just prior to the reception of a new indication code.

In a particular station period, there can be no change from one period to the next so that a storage tube which is conducting in one duty cycle can, after being momentarily extinguished by the indication clearing amplifiers 144, immediately be brought back to a conducting state by the newly indication code received. Although the corresponding storage tube is thus momentarily in a non-conductive state, the indicator lamp associated with this tube will remain lit. This is because a capacitor is connected across the primary winding for the saturation transformer. Capacitor 503 is e.g. connected in parallel with the primary winding of transformer 499 in the anode circuit of tube 494. The charge stored in such a capacitor during the time the tube is conducting is discharged during the short interval that the tube is non-conducting, with the result that the condition in relation to the saturated transformer is virtually unchanged, and the indication lamp which is controlled thus becomes active.

Fig. 10 shows the way in which the sand band system of the present invention can be used to control various devices such as signals and pens in a typical local place and also shows how different indicating relays are controlled in the conditions of the place.

The different control relays that we see in fig. 10 correspond to those shown in Fig. 7C. These relays are controlled using different steps in the shift register. A typical site is shown including a line on a main line such as a crossing; track with traffic which is selectively routed £ a pen which is affected by switching machine 1SM. Signals ILA and 1LB are arranged so that they control the movement of westbound traffic that passes through the pen 1SM 1 left and the signal IR similarly controls eastbound traffic to the right through the pen. The track section IT is anonymous to detect the presence of trains in the area and the track relay 1TR is connected to it.

The attraction of relay 1RGZ causes the clarification of the signal IR. The special visible signal shown by this sigm is controlled by local circuits at the field station location. The attraction of relay 1LGZ similarly causes the clarification of either signal ILA or 1LB in accordance with the existing traffic conditions. When both of these relays 1RGZ and 1LGZ have released, all the signals are controlled on the spot so that stop indications are displayed. Switching machine 1SM is controlled by relay 1WZ. The attraction £ relay 1WZ causes the brush to be affected and go to its normal position, while re 1WZ in its released position causes the stove to be affected to its inverted state

Relay 1RRGP is a signal indication, repeater relay. When the signal IR is controlled to display a stop instruction, re 1RRGP is attracted. For any other single indication of this signal IR, release relay 1RRGP. Relay 1LRGP repeats the condition for these signals ILA and 1LB in the following way. Track occupancy indication arranged in accordance with the permit for the track repetition relay 1TR. Nine track section IT is busy, the now painted energized track relay 1TR falls. The currently affected position of the brush with the help of the switching machine 1SM is in agreement with the states of the brush and other relays 1RWR and 1NWP. When the switching machine has been affected to its full no male state, relay 1RWR is attracted. Nine the switching machine is affected to its on-turned state, relay 1 NWP is attracted.

These different indication relays st;

rer the conditions for corresponding steps in fo

the shift register in the manner previously described. These relays will thus determine the structure of the indication code

which is sent back to the manager's office and thus determine the information displayed

in the manager's office.

Having described an electronic code communication system in a particular embodiment

of this invention, I wish it

it should be understood that various other forms may be provided to illustrate the particular embodiment of this invention without

in any way to go outside the scope of

the present invention.

Claims (4)

1. Electronic code communication system, especially for railway signaling systems, for the transmission of electrical control signals and indication signals between a manager's office (CO) and at least one field station (FS), where a transmitter device (13, 14) arranged to send successive cycles of control signals having a mutual time gap to the field station or each field station in turn during a cycle, while at the field station or each field station there is placed a transmitter device (24) arranged to send indication signals back to the management office in response to that the field station receives control signals that have been sent to the relevant field station from the management office, characterized in that the field station or each field station has a signal storage device in the form of an electronic displacement register (19), a circuit (22) for setting the steps in this register for storing a sequence of indication signals, as well as a further circuit (18) controlled by the controller nals received from the management office to tap out the stored signals from the register, so that this taping takes place timed in relation to the control signals received at the field station. 2. System in accordance with claim 1, characterized in that the further circuit (18) is designed to successively introduce the control signals received at the relevant field station in the first stage of the displacement register, thereby gradually tapping out the indication signals already stored in the register in turn, so that the last step in the register and transfer it to the transmitter device (24), so that one indication signal is sent back to the management office for each control signal received by the field station. 3. System in accordance with claim 1 or 2, characterized in that the management office contains a stepping circuit (43) which determines the time interval between the control signals and which encodes these signals in dependence on a coding pattern, and that the further circuit (18) at the field station or each field station contains an additional stepping circuit (86) designed to work synchronously with the head office's stepping circuit and to provide decoding of received control signals in order to thereby determine which of these signals relate to the field station where this additional circuit is located, and to supply only those signals which belongs to this field station to the first step in the displacement register of this field station. 4. System in accordance with claim 1, 2 or 3, characterized in that each field station contains several adjustable controlled devices (21) which over said circuit (22) selectively prepare the various steps in the shift register (19) prior to the reception of control signals intended for this station, as well as another circuit (20) for controlling the settings of the controlled devices (21) depending on the settings for corresponding steps in the shift register when the control signals intended for this station have been stored in the register.