BE485525A - - Google Patents

Description

       

   <Desc / Clms Page number 1>
 



  "ELECTRICAL SWITCHING SYSTEM"
The invention relates to electrical communications systems, that is to say to electrical systems for transmitting signals of the frequency of speech or other signals. an object of the invention is to provide an improved communication system in which time-spaced signals are employed to transmit information.



   The invention is especially suited to automatic telephone communication systems. In known systems of this type, electrically controlled mechanical devices are used to connect
2 any of a multiplicity of two-wire or four-wire conversation circuits with any of a multiplicity of other intercommunication circuits or lines, and these connections may be made simultaneously.



   The present invention, in one of its aspects, relates to the establishment of connections between one or more inter-communication lines and a single main or junction line specific to transmission.

 <Desc / Clms Page number 2>

 to put information to the intermocmunciaotn lines, and its object is to provide improved means for connecting the main line to an intercommunication line, another object of the invention is to provide improved means for connecting the any of a multiplicity of circuits with any of a multiplicity of ingermocunicaton lines,
According to the invention, a multiplex communication system comprises means for applying signals to a single circuit, a multiplicity of other circuits, each of which is suitable for being connected,

   at desired time intervals, with said single circuit, passage means and means for intermittently actuating said passage means, so as to place said single circuit in operational connection with one of said other circuits to inter. desired time values.



   According to the invention, a communication system comprises means for applying signals spaced in time to a single main or junction line, a multiplicity of intercommunication lines, a multiplicity of passage means, each of them associated. to one of said intercommunication lines and arranged. for controlling the connection of an intercommunication line with said main, or junction line, and means for intermittently actuating a passage means so that an inter-communication line is functionally connected with a main line or junction for the purpose of transmitting signals from the main line or junction to said intercom line, at the time intervals at which signals appear on the main or junction line.



   According to another characteristic of the invention, each passage means is arranged to be actuated by means which comprise a source of electrical pulses which are organized to control an electronic passage valve.



   The invention further provides a multiplex communication system which comprises a trunk or trunk line, a line

 <Desc / Clms Page number 3>

 inter-connection suitable for transmitting signals to said main line, or junction, a multiplicity of circuits, passage means associated with this inter-function line and arranged to control the connection of said inter-communication line with the main line, or junction, and means for intermittently actuating said passage means so that the intercom line is functionally connected to the main or junction line for the transmission of signals from the intercom line to said main line, or junction,

   at time intervals such that the trunk or trunk line is intermittently connected with a desired circuit *
Furthermore, in accordance with the invention, a multiple communication system comprises a main line, or trunk, a multiplicity of circuits each of which is suitable for transmitting signals spaced in time to said main line, or junction, a multiplicity of intercommunication lines, a multiplicity of transit means, each of them associated with one of said intercommunication lines and arranged to control the connection of an intercommunication line with said main line, or junction,

   and means for intermittently actuating any of said passage means so that the associated intercom line is operatively connected to a trunk or trunk line for the transmission of signals from the main line or function to said associated intercommunioation line, at the time intervals at which signals appear on one of said circuits.



   The invention further consists of a multiplex communication system which comprises means suitable for generating electrical pulses spaced apart in time, a multiplicity of calling circuits arranged to transmit audio-frequency signals, means for modulating these pulses by said audio-frequency signals, such that the signals from any of said calling circuits modulate separate pulses from one

 <Desc / Clms Page number 4>

 on the other by fixed, predetermined time intervals, a main or common junction line arranged to transmit the modulated pulses emanating from all of the said call circuits, a multiplicity of intercom lines, a multiplicity of means of passage ,

   each of them associated with one of the intercommunication lines and arranged to control the connection of an inter- communication line with said main or junction line, means for intermittently actuating each of said means of passage so that the associated intercommunication line is operatively linked to the main or junction line for the transmission of signals from the main or junction line to said associated intercommunication line, at time intervals corresponding to the appearance of pulses corresponding to the signals of one or more said calling circuits on said main or junction line, and means for modulating the pulses received by an intercommunication line, by means of to which audio-frequency signals are transmitted over the intercom line,

   
The invention further provides a multiplex communication system, in which audio-frequency signals appearing on any one of a multiplicity of preservation circuits, all adapted to be connected simultaneously with a main line or a main line. common junction, are caused to appear on the main or common junction line as? pulses spaced apart in time and are transmitted on any line chosen from a multiplicity of intercommunoation lines,

     which is connected to the main or junction line by passage means actuated to functionally connect the main or junction line with said inter-communication line at time intervals consistent with the appearance on the main line or junction of pulses corresponding to signals on a conversation circuit,
The invention is illustrated by way of example in the accompanying drawings, in which:

 <Desc / Clms Page number 5>

 
Fig. 1 is a diagram illustrating an arrangement providing selection means for connecting, via a main or common trunk line, a first multiplicity of intercommunication lines with a second multiplicity of intercommunication lines;
Fig. 2 is a diagram similar to FIG. 1, comprising a variant of selection means;

   
Figs. 3a, 3b, 3c and 3d taken together show the arrangement of an automatic telephone exchange embodying the invention;
Fig. 4 is a timing diagram illustrating the relationships between pulsations generated in the arrangement of the freezes. 3a to 3d.



   Referring to the accompanying drawings, T11 and T'il show respectively the 2-wire transmitter and receiver circuits of a 4-wire intercommunication line which is part of a first multiplicity of 4-wire intercommunication lines, while that T1 and T'12 represent the 2-wire transmitter and receiver circuits of a second 4-wire intercommunication line in the first multiplicity of 4-wire intercommunication lines, the arrows indicating the direction of transmission for each of the lines.

   When any line of the first multiplicity of 4-wire intercom lines changes from the "free" condition to the "engaged" or "busy" or interlocking condition, that is, it "calls" ", the calling line is connected, as will be described, to a 4-wire intercom line, which is free in a second multiplicity of 4-wire intercom lines, one of which includes transmitter circuits and 2-wire receiver, T21 and T'21 respectively, is shown.

   The 2-wire transmitter circuits, T11, T12 are individually connected to a time division multiplex transmission system, 100 which generates, on a single output circuit, or main or single trunk line, 101 , key signal pulses spaced in time, corresponding in time order to circuits T11, T12 ..... and so on.

   If we assume that there are one hundred circuits such as Tll, that the time division pultiplex system 100 provides one hundred trips

 <Desc / Clms Page number 6>

 which are individually intended for one of the 2-wire transmitter circuits T11, T12 that the pulse repetition frequency on each multiplex path is 10,000 per second, the pulses appearing on the trunk or trunk line 101 will be spaced apart at intervals of 1 microsecond * In the example described, arrangements have been made so that, when a 2-wire transmitter circuit is engaged, the corresponding pulses emitted by the multiplex system 100 have a finite average amplitude, currents transmission being transmitted by pulse amplitude modulation but, when the transmitter circuit is unoccupied or free,

   the pulses fall to zero or almost zero amplitude. The 2-wire receiver circuits are individually connected to a 100-path time division multiplex transmission system, 200, which distributes signals spaced in time from a single circuit or trunk or trunk line, 201, in turn. role of the circuits T'Il, T'12 ..., via low-pass filters 112, 122 ..., which demodulate the pulses at audio-frequency outputs, which are amplified by the amplifiers 113, T123 .. before being sent to the T'Il, T'12 circuits.



   The means for controlling the spacing in time of the multiplex systems 100 and 200 are coupled or locked together in such a way that the time element of the system 100 corresponding to the 2-wire transmitter circuit of a 4-way intercommunication line son coincides with the system time element 200 corresponding to the 2-wire receiver circuit of the same 4-wire intercom line. The output circuit or main line or junction 101 is assembled with a multiplicity of through circuits of which one, 102, is shown and of which one is provided for each of the 2-wire transmitter circuits of 4-wire intercommunication lines. which form a second multiplicity of 4-wire interconnection lines.

   A single 2-wire transmitter circuit is represented and designated by T21, The corresponding 2-wire receiver circuit is represented by T21 and it is connected, via a through circuit, 202, to a main line

 <Desc / Clms Page number 7>

 or input junction, common, 201. The passing circuits of other 2-wire receiver circuits are connected in a similar way to circuit 201.

   The passage 102 is opened by an impulse sent to the conductor 103 and the passage 202 is opened by an impulse sent to the conductor 203. The conductors 103 and 203 are both connected to the conductor 301 on which, as will be described, pulses are sent coincident with the path pulses of any one of the circuits of the first multiplicity of 4-wire intercom lines for which a link is required.



   The input to a 2-wire circuit, for example T11 in the transmitting direction and at T21 in the receiving direction, includes a DC signal which is amplitude modulated by talking AC and other signals. . In the absence of an input, the path pulses in the corresponding multiplex have practically zero amplitude.



   The DC component of an input causes the path pulses to assume a standard amplitude, which is modulated by the AC component of the input.



   Demodulation of the pulses by low pass filter 400 produces a modulated DC signal which is a replica of the input except that it is greatly attenuated.



     Amplifier 500 could amplify both components of direct current and alternating current, but in practice this is not desirable. Amplifier 500 therefore amplifies the AC component which it supplies to the 2-wire path T21. A suitable device demodulates the pulses and amplifies the DC component, the DC and AC outputs being recombined at the output T21. In the reverse direction of transmission, the same arrangement is used to demodulate and amplify the pulses coming from the multiplex 200, the low pass filters and the amplifiers handling the ac components and other devices the components. direct current.

 <Desc / Clms Page number 8>

 
 EMI8.1
 



  -.! '. .v r 'CI. - A line detector comprises a trigger circuit 300 which, in itself, is of the well known type, comprising two positions of electrical stability. This deolanching circuit is fixed in one or the other of its two stability positions by pulses. Circuit 300 is brought from its first position to its second position by a pulse sent to conductors 302 or 303. It is brought from its second position to its first position by a pulse sent to conductor 310. This conductor 310 is connected. to a source of pulses K2 which are separated from each other by one microsecond and shifted by a fraction, for example 1 / 5th, of a microsecond in advance of the multiplex pulses of the main line or common junction 101.

   The trigger circuit 300 while in its second position, also causes a pulse, preferably alternating current, è, for example, 15 megacycles per second, to be emitted by conductor 304 to a line. electric or supersonic delayed 305, from which the pulse will emerge, in the presently described example after a delay of 99.6 microseconds plus or minus a tolerance of, for example, 1 / 5th of a microsecond. The pulse of the delay line is sent to a pass circuit 307 through the conductor 30b.



   In connection with the circuit 307 passing through the conductor 308, there is provided a source of pulses K1 each of which has a duration of, for example, 1 / 5th of a microsecond, and which are spaced apart by one microsecond, while that they practically coincide with the path pulses.



   A K1 pulse will occur approximately in the middle of the duration of a pulse received by conductor 30b and, during pulse K1, a pulse is emitted and sent to conductor 303 to cause circuit 300 to drop out of its first step. in its second position and thus give rise to other pulses emanating from said trigger circuit and sent to conductor 304 as already described.



   The process by which the trigger circuit 300 is brought from one of its positions to the other is thus repeated. The trigger circuit 300, the delay line 305 and the through circuit 30 '(, with their associated circuits, thus constitute a

 <Desc / Clms Page number 9>

 circulation which, when started, produces and sends a pulse on conductor 301 every microsecond cemetn to act on passages 102 and 202.



   The K1 pulses of conductor 308 are synchronized with the multiplex systems 100 and 200, so that the passages 102 and 202 are opened as described at times which coincide with the occurrence of the pulses of a path in each of the multiplex systems. .



  The pulses passing through passage 102 are sent through low pass filter 400 to be demodulated to audio-frequency, which is amplified by amplifier 500 and sent to intercommunication line T21 as already described. Audio-frequency currents entering the intercommunication line T'21 are modulated into spaced pulses by the passage circuit 202 which is open for the pulses appearing on the conductor 203 and the modulated pulses are put in. common with the modulated pulses coming from other intercom lines entering on the main line or trunk 201.

   It therefore appears that any pair of circuits, for example T11, T'll, in the first multiplicity of 4-wire intercom lines, can be electrically connected to any of the 4-wire intercom lines, for example T21, T'21 ... in the second multiplicity of 4-wire intercommunication lines provided that the circulation system described is arranged to emit on the conductor 301 pulses corresponding in time with any of the paths multiplex chosen in systems 100 and 200.



   When no circuit of the first multiplicity is engaged or occupied, all the pulses sent to the conductor 101 will then have zero amplitude. If now one of the circuits T11, T12 is to be engaged, pulses will appear on the trunk or trunk line 101 in a situation or time psion relation which corresponds to the busy circuit.

   It has already been indicated that the duration and the spacing of the pulses have prescribed dimensions and it will be understood that the time which elapses between the im-

 <Desc / Clms Page number 10>

 pulses which correspond to one path is occupied by pulses which correspond to the various other paths so that, if more or one circuit is engaged / occupied at any given time, a number of pulses will appear simultaneously on the main or junction 101, these pulses being constituted by the pulses which have been described and which appear one after the other on the main or junction line. For the presentation only a single pulse will be considered, that is to say the pulse which appears when a circuit T11, T12 ... comes to be engaged or occupied.

   The main or junction line 101 is connected by the conductor 110 to a passage circuit 700 which is normally open in order to allow the pulsations sent on the conductor 110 to pass on the conductor 7ol and beyond, via the conductors 602, to through circuits 600, one of which corresponds to each intercommunication line such as T21, T'21
The pass circuit 700 is closed by pulses which are received on line 702, these pulses being supplied to line 702 on delay lines 305 which are individually associated with intercommunication lines T21, T'21 ...

   The transit circuit
600 allows pulses received on the common conductor 701 and the conductor 602, to pass to the trigger circuit 300 through the conductor 302, only when a marking signal from a line marker is applied to the switching circuit. passing through the conductor 601. The line marker is made suitable, by means not shown, to mark only one passage circuit 600 at a time, and only one passage circuit which is associated with an intercommunication line free in the second multiplicity of 4-wire lines.

   The first impulse which must appear on the main line or common jenction 101 when it is necessary to call or connect one of the lines T21 ..., passes on the conductor 110, through the passage 700 which is open at this moment, through a passage 600 marked by the line marker, on the conductor 302 and thus bring the trigger circuit 300 from its first to its second position, thus determining the start of the circulation system as described above.

 <Desc / Clms Page number 11>

 



   The next and subsequent pulses which are to appear on the main or ICI function line from the calling circuit, coincide with the pulses emitted on conductor 301 and hence electrically connect the calling line to the marked output line.



   Said next and subsequent pulses are prevented from passing through passage 700 by the Pulses which appear to coincide with them by conductor 702 from conductor 309 and hence cannot cause another circulating circuit to begin. emission of generating pulses. When the connection has been made as described, the demodulation direct current component obtained from a demodulator apparatus is sent by a conductor to the line marker to indicate that the marked pass-through line and its line detector have been. commissioned.

   The line marker then marks another free intercom line, which will connect itself as described to the next incoming line which is to be engaged or connected. As the disappearance of the marking through line 601 is relatively slow, if two circuits call and start their path pulses together in a very short time interval, there is the risk of connecting the soft calling circuits to the same line. of intercommunication.

   For this reason, the pulses emitted from the trigger circuit 300 are sent by a conductor to a device which transforms the pulses into a continuous signal applied to the passage 600 so as to close this passage as long as the signal persists, and this continuous signal is applied in advance of the disappearance of the marking effected by the line marker. The fact that the line detector has answered a calling line is signaled to the line marker by the direct current signal derived from the demodulator apparatus, and sent to the line marker which, after that, marks a new one. line detector if there is one available.



   When an incoming or calling circuit is released, the pulses on the corresponding multiplex path cancel each other out, and therefore

 <Desc / Clms Page number 12>

 The output pulses of passage 102 and the output of the demodulator apparatus also become zero. The output lead of the demodulator apparatus is connected to the apparatus 800 which is conditioned to emit a long pulse on the lead 801 when the pulses of the passage 102 and the rate of the demodulator apparatus cease.

   The pulse on line 801 maintains trigger circuit 500 in its first position despite the pulses which appear on conductor 303 -and which, in the absence of the pulse-
 EMI12.1
 Zion passing on the conductor 801 would place the circuit to start z0 in its second position.

   The circulation system is thus interrupted and the connection is broken.
 EMI12.2
 The arrangement shown in f ig 2 differs from that shown in fig.l, in that the conductor 110 is not connected to the conduc-
 EMI12.3
 turn HERE but to a line marker, and that device 00 is not connected to the demodulator device but to a control conductor to which conductor 301 is also connected instead of to a line marker.



     The arrangement in fig. 2 functions to connect a line of in-
 EMI12.4
 4-wire communication, T1, T'21 which has been marked by a signal coming on said control conductor of a previous switching stage, to the selected circuits of a wire intorcommunication line such as, for example, circuits T11, T'Il .... a line marker, not shown, is caused to emit a pulse on conductor 110 which coincides with the corresponding time element.
 EMI12.5
 dant to circuits T11, T'Il ...., e; thus to start up the circulation system associated with the intercommunication lines T21, Te at the desired time.

   The desired connection is then established, and it is broken when the signal on the control conductor disappears.

 <Desc / Clms Page number 13>

 



   Referring to figs. 3a, 3b, 3 ± and 3d and 4, figs. 3a to 3d show the circuit arrangements for a simple automatic telephone exchange with 10 lines, which would be suitable as a private exchange serving subscribers not far from each other, for example located in the same building.



   A subscriber equipment. is shown as constituted by the lines' 1 and L, together with the apparatus shown between these lines and the tubes or valves 1 and 11 of FIG. 3a. A second subscriber line is similarly arranged and connected to valves 2 and 12, but has been omitted from the drawings. The valves 3 and 13, 4 and 14., 10 and 20 are also connected to subscriber lines by similar devices. At each subscriber station is placed a conventional hand-held device consisting of a transmitter of the carbon pellet type and an electromagnetic receiver, a hook-switch to establish direct current conditions of call and end of conversation, and an impulse dial-disc, all connected to a pair or T circuit.

   In addition, each subscriber is provided with a signaling device, not shown, for example a lamp and a battery connected to the pair L,
The valves 1 to 10 are arranged to constitute pulse modulators and serve, together with the associated apparatus which will be described, as a multiplex device for converting, in a time division multiplex process, the signals into direct current and audio. -frequency, which appear on pairs such as T, as signals spaced in time, amplitude modulated, on a single output conductor, that is to say the common anode circuit of the valves.

   After polarity reversal by a valve 31 and power amplification by a valve 33 with cathode load amplifier assembly, impedance balancer, the time signals are sent, on the conductor G, to a multiplicity of first selectors, or as well / they will henceforth be designated as line detectors, the circuit of one of these line detectors being shown in FIG. 3c.

 <Desc / Clms Page number 14>

 



   Each line detector, as shown in fig. 3c is directly connected to a selector; there are therefore a number of such selectors and the circuit diagram of such a selector is shown in fig. 3d.



   The valves 11 to 20 shown in fig. 3a provide the means by which speech or signals received from distant subscribers, which will appear on conductor R as time-spaced, amplitude modulated signals, are distributed to subscriber circuits.



   Each line detector is associated with, in addition to means allowing it to be connected to a calling subscriber by the action of the line detector, a device designed to answer the number of the desired subscriber and also establish the call. connection.



   The marking of an unoccupied line detector, and consequently the marking of a free intercommunication line, is carried out by a line marker, the circuit of which is shown in figure 3c, to the right of the vertical line in broken lines, both the line detectors and the selectors are connected to the common multiplex conductors C and R, via through valves such as% that the valves 41 and 42 shown in FIG. 3c.



   As already described, the information from a calling circuit, the subscriber circuit in the arrangement which is now described, is transformed into a pulse which consists of pulses evenly spaced in time.



   The interval between the midpoints of neighboring pulses of one of these pulses can be called the multiplex period, the reciprocal of the period being the repetition rate of the pulse, and the time during which a pulse can persist is Referred to as path spacing, the term path refers to a path through a multiplex device.

   The multiplex frequency must exoerate twice the highest audio frequency used to modulate the pulses in order to obtain satisfactory transmission,

 <Desc / Clms Page number 15>

 
In the accompanying drawings, the screen grids of pentode valves are shown disconnected and it should be understood that the conductors leading to these grids have been omitted to avoid confusion, and that they should be connected in the normal manner to a source. of potential that is suitable for the operating conditions and characteristics of the valves.



   As indicated above, ten paths are provided in the exchange shown in the drawings. The multiplex pulse repetition rate used is 10,000 cycles per second, so each path can theoretically be connected to the main or common trunk line G for a 10 microsecond path period during each 100 microsecond multiplex period * In practice, however, because of the need to ensure that each path is certain to be disconnected before the next is connected to the main or common trunk line, and because of the limited speeds at which switching operations can be performed It is advantageous to arrange for each path to be connected to the main or common trunk line for only a fraction of the path spacing.



   A suitable fraction is 1/5 and in the description which follows it will be assumed that the duration of a pulse corresponding to a path is 2 microseconds.



   The valves 1 and 11, 2 and 12 ... etc, are cut, that is to say made non-conductive at their anodes by a bias voltage V5 which is applied, via resistors such as R6, to the valve stop grids. The voltage V5 is sufficiently negative with respect to the supply voltage V3 of the cathode, to suppress the anode current. The valves can be made active or conductive by, and for the duration of pulses going in the positive direction and hereinafter called positive direction of a value substantially equal to the difference between the voltages V3 and V5. applied to stop grids via capacitors such as C6.

 <Desc / Clms Page number 16>

 



   Positive direction pulses applied to capacitors such as C6 come from a pulse generator PG. This generator can be set up in accordance with well known practice or technique, and any suitable arrangement can be employed. The generator must generate pulses on twelve output conductors marked K1, K2, P1, P2. P10.



   The pulses generated are illustrated by the timing diagram of fig. 4.



   The conductor K1 carries a continuous pulsation which will be called the pulsation K1, and which consists of a positive sense pulse of almost rectangular shape, each persisting for two microseconds and occurring with a repetition frequency of 100,000 cycles per second. This K1 pulse is the primary means of adjusting system times and can be achieved by the well-known method of squaring the sinusoidal output of a 100,000 cycles per second oscillator, differentiating and resetting to obtain the waveform. asymmetric desired. The conductor K2 receives the pulsation K2 which consists of pulses going in the negative direction and hereinafter referred to as negative direction, of the same shape, duration and repetition frequency as the pulsation K1 but organized to occur approximately eight microseconds later.

   It can be obtained by using the same source of sine waves as that used for the pulsation K1, the said waves being passed through an appropriate phase transformer network before being subjected to the process of squaring, differentiation, re-squaring. and reverse polarity, necessary to give the required waveform.



  The conductor P1 receives the pulsation P1 which can be obtained by passing each tenth pulse from K1 to the conductor P1. This can be achieved by using a source of sine waves with a frequency of 10,000 cycles per second, synchronized with the oscillator at 100,000 cycles per second, the waves being suitably phased, offset, differentiated and re. - squared to allow action on a passage amplifier located between the conductor K1

 <Desc / Clms Page number 17>

 and the conductor P1. Such a passage amplifier can be a pentode valve arranged in a known manner so as to be conductive only at desired times.



   Another method of producing the pulsation P1 can consist in the use of an electronic two-quinary counter supplied at the input by the pulsation K1, and emitting at its output a quadrangular wave which, after differentiation, can be used to act on a passage amplifier located between conductor K1 and conductor P1.



   The individual pulses on the conductor P1 are thus synchronized with, and are of similar polarity, duration and shape, as one and every tenth pulse K1. The conductor P2 receives the pulse P2 which is from all points of view analogous to the pulse P1 except that the pulses P2. are delayed by 10 microseconds compared to the pulses P1. The P2 pulsation can be generated as in the first method recommended for the P1 pulsation but with an appropriate phase changer circuit inserted in the control circuit at 10,000 cycles per second. In a similar way the pulsations P3 ....

   P10 are supplied to the appropriate conductors, each of which is obtained in time relation with respect to the pulse of lower number as recommended above for obtaining the pulsation P2 in relation to the pulsation P1.



   As shown in fig. 3a, the anodes of valves 1 to 10 are connected to a common anode resistor R3. Resistor R4 and capacitor C4 are provided to decouple the supply voltage V1 which is advantageously positive with respect to V3.



   When subscribers are unoccupied or free, their circuits T are open and under these conditions the control gates of valves 1 to 11 are at earth potential. The cathodes of these valves are connected, via resistors such as R5. to a supply voltage V3, positive with respect to the earth, such that all the valves are cut off at their control gates. Thus, despite the pulses which are supplied from the generator PG to the stop gates

 <Desc / Clms Page number 18>

 valves, no pulse is emitted in the common anode circuit when the paths are unoccupied or free.

   If a subscriber engages his line, direct current flows from the positive terminal of battery B of the exchange, via one half of the primary winding of the hybrid transformer T1, one branch of the pair T the transmitter contacts, dial and contact switch the subscriber, back along the other branch of pair T, the other half of the primary winding of transformer Tl, and through resistor R1 to the earth terminal of the central unit battery.



   One effect of said supply current is to raise the potential of the junction of R1 and of the primary winding of the transformer T1, from that of the earth to the value at which the rectifying element W1 conducts and thus stops the potential rise to a very slightly positive value with respect to the DC supply voltage V3. This rise in potential is communicated, via resistor R2 to the control grid of valve 1, and the value chosen for cathode resistance R is such that, in this condition, valve 1 functions satisfactorily as an amplifier. . Valves 2-10 are analogously controlled or controlled by the engaging action of the lines associated with them. When, therefore,

   any subscriber engages his line, a negative sense path pulse is developed across common anode resistor R3. This path pulsation is developed by, and is therefore synchronous with, the pulsation P1 to P10 applied to the stop gate of the particular valve 10 reserved for this subscriber.



   This characteristic of the invention constitutes the call and hold signal of a calling subscriber and, as will be seen, the response signal of a called subscriber.



   Similarly, when other subscribers engage their lines, they cause the generation of path pulses across resistor R3 which, polarity and amplitude apart, are the replicas of pulses P1 through P10 which are reserved for them.

 <Desc / Clms Page number 19>

 



   When a subscriber disengages his line, the voltage at the control grid of the associated valve 1 to 10 drops to earth potential, shutting off the valve and thus eliminating the associated path pulsation of those, if he there are, appearing across the resistor R3. This characteristic constitutes the release or end of conversation signal.



   The magnitude of any particular path pulsation depends on the control gate voltage of the associated valve 1-10.
In the absence of any potential drop across resistor R2, the path pulsation will have a constant amplitude which will be called the standard amplitude. Words or signals entering the pair T and developing alternating currents in the primary windings of transformer T1, will generate an alternating voltage in the secondary windings of T1. Part of this voltage is, via capacitor C2, developed across resistor R2, and thus will amplitude modulate the path pulses emitted from the anode of the associated valve 1 to 10. This is the means by which the speech or alternating current signals are transmitted through the multiplex.



   The dial or number pulses are transmitted in the same way as the call and release or end of call signals, namely: each time the circuit of a subscriber line is opened, the amplitude of the pulses of the associated path pulsation cancel out and when the line is closed the K pulses begin.



   It should be observed that time means are provided on the selector to differentiate the dial and release signals.



   Path pulses appearing across Resol reversed in polarity by amplifier 31 and then communicated, via capacitors C7 and C8, to the control grids of cathode-loaded tubes or valves 32 and 33. Valves 32 and 33 have their own. anodes connected to the potential source V1 and their cathodes connected to earth through cathode resistors.

   The grid of con-

 <Desc / Clms Page number 20>

 The control gate of valve 32 is biased, via resistors R7 and Rg, to the direct current supply voltage V, and the control grid of valve 33 is biased via R8 to the same voltage V4, which is made negative with respect to earth to an extent sufficient to keep valves 32 and 33 close to the cutoff point, thereby allowing them to act satisfactorily for the positive path pulses.



   The cathode-charged valve, 32, impulses to lead 0 and the cathode-charged valve 33 similarly supplies lead G.



   Until now only the transmission of the conversation and incoming signals from the subscriber has been considered and it has been explained that this conversation and these signals appear on the main line or G junction in the form of amplitude modulations. The conversation or signals transmitted to the subscriber appear on the trunk line or R junction as amplitude modulations superimposed on standard amplitude path pulses superimposed on the standard amplitude of the path pulses .



   As can be seen, these pulses are formed and modulated in the line detector or selector to which the subscriber is connected and are synchronous with the path pulses on the main line or trunk to go G. On the main line or from junction, R, appear impulses of negative direction; these pulses are reversed in polarity by amplifier 34 and communicated to the common control grills of valves 11 to 20.

   These valves have their cathodes connected, each by a cathode resistance, to the supply voltage V3, and the control gates connected to earth by a common grid leakage resistor, the cathode resistance being chosen in such a way by compared to the characteristics of the valves, to the voltage V3 and to the standard amplitude of the positive direction path pulses, that each path pulse raises the potential of the control gates in the range of active potentials, in which the valves amplify in a way that satisfies

 <Desc / Clms Page number 21>

 health.

   The valves 11 to 20 have their stop grids connected to those of the valves 1 to 10 respectively, and are therefore capable of the same operation, namely to be cyclically put into operation for the duration of the pulses supplied on the conductors P1. to P10.



   As, as said, the appearances on the main or junction lines G and R of any particular path pulse are synchronous, it follows that the path pulse on the main line or junction R will be admitted to passage through the suitable valve 11 to 20 and not through any other.



   As will be described later, the trunk or trunk R path pulse does not appear until the called party answers. When this occurs, negative direction path pulses appear at the anode of the relevant one of valves 11 to 20, and connector C9 and the control gate of valve 35 are rapidly negatively charged, via diode D1. , sufficient to cut the valve 35 and release the normally actuated relay S which, via the contacts S1, completes the circuit between the branches of the pair L. This constitutes the response signal already mentioned.



   The anodes of valves 11 to 20 are also each connected to a low pass filter 112 'having a cutoff frequency of slightly less than 5,000 cycles per second.



   Signals of the voice frequency existing as modulating on the path pulses pass the filters and are then amplified by valves 113 'and transmitted, via transformers T2 and T1 to the subscriber line. The purpose of amplifier 113 'is to compensate for losses resulting from, among other things, the multipelx The loss from / multiplex arrangement of a ten path system, using rectangular pulses of only 1 / 5th of the travel period., is 49/50 of the incoming power.



   The arrangement of the windings of transformer T1 will be well understood by those accustomed to telephone practice. The condensa-

 <Desc / Clms Page number 22>

 tor C1 is set large enough to provide negligible impedance at the voice frequencies compared to that offered by the primary windings of T1 and ZB in a balanced network. It should be understood that the system has obvious inherent possibilities for amplification which, although requiring precise ZB balancing impedance, can provide possibilities for longer or less expensive local lines.



   A number of line detectors are provided, equal to the number of simultaneous conversations required, and arranged as shown in FIG. 3c. The common conductors, to which all line detectors are directly connected, are, as shown, on the left side of the diagram. Above, to the right of the broken line in fig. 3c, the circuit of the line marker LFA is shown, the latter comprising an automatically controlled single selector SA of a well-known type.



     The uni selector SA has two sweepers, each of which works on a series of contacts. One sweeper is arranged to be earthed and the other, via the control magnet, SA, and the associated mechanically actuated switch contacts, SA d.m., to the central battery. The LFA line marker is connected to each line detector by two conductors, such as W and X, which are connected to corresponding contacts in both series.



   The line marker is adjusted in such a way that if the M series contact on which the SA unit is stopped becomes grounded, the sweepers will automatically move until a non-ground contact is found.



   The W conductor is connected to form the E1 contact of the E relay, the other side of the contact being earthed. As will be understood, when a selector is engaged, the relay E is actuated, thus preventing the single selector SA from remaining on the contact W corresponding thereto.



   As a result, the line marker scanners will always be on the X and W contacts of a non-line detector.

 <Desc / Clms Page number 23>

 engaged, assuming a line detector in this condition.



  In this state the line marker is said to mark this line detector to take the next call, a function performed by extending the potential, from ground on conductor X to resistors R43 and R44 of the line detector. When this last line detector comes to be engaged, the scanners move to mark another unengaged line detector.



   In fig. 3c the valves 41 and 43, and 42 and 44 are the main line or junction passages by means of which line detectors are connected to the main or junction lines G and R respectively, that is - that is to say constitute the passages 102 and 202 of FIG. 1. When a detector is not engaged these passages remain closed, that is to say that the valves 41 and 42 are rendered inoperative.



   When a line detector becomes engaged, these valves are cyclically made active, coincident with the occurrence on the main or junction line G of the engagement path pulse. The preferred timing device for cyclically actuating trunk or trunk crossings comprises a circulating system in which a single pulse, of the path pulse to be selected, can be introduced, and which contains a delay device having a delay. approximately equal to the multiplex period.

   The circulation system is arranged in such a way that, once started, pulses will circulate until the moment they are deliberately stopped, and at each passage through the point of entry into the System, will produce the required operation of the devices. main line or junction crossings.



   The delay device employed in the circuit of FIG. 3 is a supersonic mercury-filled delay line, DL, of the type which has become familiar to technicians in many fields, and in particular to those dealing with counting or calculating systems.



   Several such devices are described in an article entitled "An Ultrasonic Memory Unit for the E.D.S.A.C." by M.V.



  Wilkes and W. Renwick, published in the periodical "Electronie Engineering"

 <Desc / Clms Page number 24>

 Vol 20, No 245 July 1948 and in an article titled "Mercury Time Lines" by T. K. Sharpless, in the periodical "Electronics" dated November 1947, at page 134.



   Those skilled in the art will understand that practical difficulties arise when seeking to construct a multiplicity of time units, such as a mercury-filled delay line, each having such stability with respect to the main time device, namely the K1 pulse, that once switched on these units will remain exactly synchronized with the K1 pulse for an extended period.



   It is possible, however, and means will be described for, to overcome these difficulties by employing a source of synchronizing pulses and using them to rephase the circulating pulsation each time it reaches the point of introduction and. in this way thus preventing the cumulative effect of the error in the delay line, of limiting the error to that of a single transmission through the line. Under these conditions, the delay stability of the mercury delay line is suitable for the intended purpose.



   The valves 40 and 4 of FIG. 3c are the pulse introduction and re-phasing passages respectively. A negative sense pulse emitted from the common anode circuit of these valves, is organized to be communicated to the gate of the normally conductive valve 46, thereby changing the unwinding valves 46 and 47 of the valve. first to second position of stability * The anode circuits of the trigger valves 46 and 47 are connected to a balanced solid state relay comprising the rectifier elements W4, W5, W6 and W7 and associated transformers and resistors, in such a manner that when the triggering device is in the first stable position,

   the attenuation between a carrier frequency oscillator CG and an amplifying valve 48 is very large and, when the trigger device is in the second position of stability, said attenuation is made small. Consequently, following the change of the trigger device,

 <Desc / Clms Page number 25>

 the carrier frequency is admitted to the amplifier 48 until the moment when the triggering device is replaced. This occurs 8 microseconds later due to the incidence at the stop gate of valve 47 of the negative direction pulsation K2 developed through resistor R48 and supplied via capacitor 041 from of the pulse generator PG.



   The 8 microsecond carrier frequency pulse is amplified at valve 48 and transmitted by the output transformer of this valve to the transmitting quartz plate of the retetedmet line,
The delay line is set to have a delay of 96 microseconds, and at the end of this time the carrier frequency pulse emerges, greatly attenuated, from the receiving quartz plate.



   After passing through the delay line, the pulsation is amplified by valves 49 and 50, and part of the power rectified to diodes D4 and D5, to produce a pulsation of positive direction across resistor R48, sufficient to bring the control grid of the valve 45 to the cutting voltage to which it is biased by means of the potentiometers R49 and R50, substantially at the potential of the earth. The valve 45 has its cathode grounded and the stop gate continuously pulsed, through the capacitor C42, by pulses K1 of amplitude substantially equal to the cutoff bias voltage K5 which is applied to the gate of valve stop / 45 via R51.

   Like the delay in the supersonic line
DL is 96 microseconds, it follows that valve 45 will be activated, at the control grid, before the entry edge of an individual pulse K1 reaches the stop grid, and cut again at the control grid after or behind said entry edge.



   A single pulse K1 is therefore passed through the anode circuit of valve 45 and develops a negative sense pulse across resistor R40 exactly 100 microseconds after the previous pulse across the resistor.



   Thus the trigger valves 46 and 47 are again inverted and then re-established by a pulse K2 and, the process continues,

 <Desc / Clms Page number 26>

   As above, it will be observed that the circulation having / been initiated, will continue until such time as it is deliberately interrupted provided that the delay of the delay line remains in the range 92-100 microseconds.



   The negative sense pulsation which is thus developed at the common anode resistor R40, being developed first by a path pulse applied to the control grid of the valve 40 and subsequently by pulses passed through the valve 45 from the pulse K1 to the stop gate of the valve 4, is necessarily synchronous with, and lasts for the same period as the path pulsation which corresponds to the particular child who has engaged his line.



   This negative sense pulsation is used to open main line or trunk crossings to feed into the line detector the correct path pulsation on the main line or G junction and output from the line detector. a synchronous path pulsation in the trunk or trunk line R. The trunk or trunk lines include the pairs of valves 41 and 43, 42 and 44. Valves 41 and 42 are amplifiers, valve 41 having a control gate circuit connected with the main or common junction line G and an individual anode circuit, while the valve 42 has an individual control gate circuit and an anode circuit connected to the main line. blade or common junction R.

   In both cases, the anode supply voltage is earth, while the cathode supply voltage V5 is negative with respect to earth.



   Without an incoming signal and the valve 43 removed, the control grid of the valve 41 is palarized to the cut-off voltage using the voltage V6, see fig. 3a, and the cathode resistor R47 is chosen such that a positive direction path pulse, of standard amplitude, raises and passes the voltage of the control grid in the range of potentials for which the valve 41 amplifies satisfactorily.

 <Desc / Clms Page number 27>

 



   The control gate of the valve 42 is also normally polarized at the cut-off voltage, the potential on the terrestrial phantom of the return pair being, under these conditions, such, in conjunction with the voltage V8 and the potentiometer R61. and R62, that it generates the necessary polarization via the secondary winding of transformer T5.



  When, as will be noted, the called subscriber answers, the potential of the terrestrial phantom of the SI pair! is brought to a level sufficient to give certainty that rectifier W8 is conducting and thus sets the bias of the control gate of valve 42 to voltage V5, a state in which valve 42 amplifies satisfactorily. In addition to the switching operations performed at valves 41 and 42 by the changes in the control gate polarization described above, both valves are normally made non-conductive by the cathode polarization effect of valves 43 and 44. These valves have their anodes connected directly to the supply voltage V1 and their control gates polarized via the resistor R46 and the supply voltage V7 which is positive with respect to V5.

   The cathodes of valves 43 and 44 are connected to the cathodes of valves 41 and 42 respectively. The bias voltage V7 is such that, in the absence of any superimposed voltage thereon, the cathodes of valves 43 and 44 and therefore those of valves 41 and 42 are kept positive to such an extent, with respect to V5, that the valve 41 is kept off even though the path pulses of the maximum amplitude which can be expected arrive at the control grid of this valve.

   When, however, a line detector is engaged and therefore develops across resistor R40 a negative sense pulse synchronous with the path pulse to be selected, each pulse of that negative sense pulse is superimposed, via capacitor C47, on the bias applied to the control gates of valves 43 and 44, which are therefore cut off, thereby allowing valves 41 and 42 to amplify during the duration of the pulse. These are the means, according to the invention, by which the selection of a

 <Desc / Clms Page number 28>

 desired route.



   The selected path pulse which has passed through valve 41 is then sent through a low pass filter 400 and an amplifier 500 similar in construction to those 112 'and 1131 described in relation to the subscriber center, and the frequencies voices thus extracted are transmitted via the / 14 transformer to the SG pair.



   Another feature is the means by which the circulation of the pulsation is initiated in the line detector.



  Referring to fig. 3a, it will be observed that path pulses, after polarity reversal at valve 31, are supplied to the control gates of the two cathode load amplifier mount valves, 32 and 33. Valve 32 applies signals to the main line. or call junction, conductor C, and it will be observed that although normally polarized near the cut-off by voltage V4, via resistors R9 and R7, polarization for which valve 32 satisfactorily amplifies the path pulsation , the application of an alternating current pulsation to the pair PS will have the effect, after amplification of this pulsation at the transformer T6 and valve 37 and rectification at the transformer T3 and diodes D2 and D3. to superimpose a pulse of negative direction on the polarization of the valve 32.

   The amplitude of the negative sense pulse thus obtained is sufficient to render the valve 32 non-conductive even when supplied with a path pulse of the maximum amplitude which can be expected. The PS pair is common to all the line detectors which communicate to it, via the decoupling resistors such as R63 and R64 shown in fig. 3c. part of the energy of the 8 microsecond carrier frequency pulse that emerges from the amplifier following the mercury delay line.



   Thus, the passage valve 32 cannot pass to the main or call junction line C a path pulse which is already circulating in a line detector.

 <Desc / Clms Page number 29>

 



   It now emerges that when any subscriber initiates a call, the path pulse which is reserved for him will appear on the main or call trunk line, C, until the moment when the subscriber is connected to a detector. line at which time, by virtue of the means described above, said path pulse will be removed from the main or call trunk line. Referring again to Fig. 3c, the main or call junction path C is connected, via capacitors such as C40, to the control gates of valves such as valve 40 in each line detector.

   Valve 40 provides the means by which a pulse occurring on the main or call trunk path can be fed into the line detector circulation system marked by the LFA line marker. As has been said, the valve 40 has its anode connected, via the anode resistor R40, to the supply voltage V1, and the cathode connected to a socket in the potentiometer comprising the resistors R43 and R44 and R45. , the said potentiometer being connected between the supply voltages V1 and V. A second socket in the potentiometer is connected to the contact X of the line marker, the voltages V1 and V5 and the resistances of the potentiometer being chosen in a manner such as when a first selector is not marked by the row marker,

   the cathode of the valve 40 is maintained at a positive potential to such an extent relative to the control grid that the valve 40 is cut off so that it is not influenced by the maximum amplitude of the 'path pulse that can appear via capacitor C40. and when a line detector is marked by the line marker, the cathode of the valve 40 is brought to a voltage, with respect to the control grid, such that the valve 40 is then at the point of cutoff. able to respond to a positive direction pulse applied via capacitor C40.



  This voltage of the cathode of the valve 40 under the tagging condition is stabilized at the supply voltage V3 by the rectifier element W2 which conducts when the voltage of the conductor X is maintained at the potential of the earth.

 <Desc / Clms Page number 30>

 



   The control gate of valve 40 is normally biased to earth potential, via resistors R41 and R42, a state which, however, is immediately changed after introduction of the pulse into the line detector, because a proportion of the energy of the carrier frequency pulse transmitted to the mercury delay line DL, obtained through the potentiometer R52 and R53 is rectified by the diode Dg and, therefore, a voltage of negative direction is rapidly developed across. of resistor R42 by reverse biasing the control gate of valve 40 enough to prevent the introduction of a second path pulse, created by a second calling subscriber, before the LFA line marker has moved and thus ceased to mark the selector.

   The resistor R42 is shunted by a capacitor C43 which, in addition to providing the flattening or absorption of the ripple for the rectifier D9 is set so widely that the voltage negativated across R42 decreases by a small amount. only between successive carrier frequency pulses, thus keeping valve 40 off at the control gate while the line detector is held.



   This feature prevents the introduction into a line detector of pulses belonging to more than one path pulse.



   The capacitor C48 connected between the cathode of the valve 40 and earth is provided to decouple the cathode resistance during the period of introduction of a pulse.



   Another feature of the arrangement is the transmission, by the line detector, of call, busy and release or end of conversation signals. When a line detector is connected to a calling subscriber, as described above, the negative sense voltage developed across R42, in addition to closing the introduction passage, i.e. the valve 40, is communicated to the control grid of the normally conductive valve 51 via the capacitor C44. The anode of the valve 51 is connected to the supply voltage V1 via the resistor r63; the grid of con-

 <Desc / Clms Page number 31>

 trole of this valve is connected to earth via resistor R54 and the cathode is directly earthed.

   The negative direction voltage communicated via the capacitor C44 is sufficient to cut off the valve 51 and, the time constant is such at the control grid of the valve 51, that the negative voltage there decreases, but slowly and is, in fact, stopped in its decrease by the action of the path pulsation passed by the valve 41 continuously discharging the capacitor C44 via the diode D7 and the resistor D55. Resistor R55 is provided to limit the current draw by diode D7 to a value insufficient to cause appreciable distortion of the path pul sat ion passed through valve 41.

   The valve 51 is therefore quickly rendered non-conductive when the selector comes to be engaged, and it is maintained in this condition by the presence of the path pulsation passed through the valve 41. If the calling subscriber comes to cease the call by opening the circuit of its line, the past path pulsation will disappear and, the valve 41 subsequently remaining non-conductive, the valve 51 will become conductive after a period of time which depends on the time constant of the circuit of the control grid of the valve 51.



   This time constant is set such that the valve 51 will not be made conductive during the temporary absence of the selected path pulse, determined by the standard interruptions in a dial signal, but will be made conductive if said pulses remain. absent for a significantly longer period, thereby identifying the opening condition. The rise of the anode voltage of the valve 51 when the line detector comes to be engaged, is used to advance on the pair SG a voltage of positive direction, obtained via the potentiometer R58 and R59 which is connected between l the anode of the valve 51 and the supply voltage V5; this positive direction voltage causes, as will be understood, the application of a call signal to the line of the called subscriber.

   In addition, said voltage rise is communicated, via capacitor C16, to the control gate potentiometer of the valve to

 <Desc / Clms Page number 32>

 triggering 47, where it is absorbed without disturbing the triggering, by the rectifier element W3. When the calling subscriber eventually cuts, and the valve 51 becomes conductive, the resulting negative direction voltage which develops at the anode of the valve 51 is communicated to the control grid of the valve 47 and there keeps the valve cut off.
47 for a period longer than the multiplex period and, thus interrupting the circulation, releases the line detector.



   Another feature of the line detector is to provide means for the detection of dial pulses. the anode of the valve 41 is connected, via a resistor R56, a diode D8 by means of which, when the line detector is engaged, a negative voltage is developed on the control grid of the valve 52, - sufficient to make this valve non-conductive, so that the normally actuated relay A is released. The cathode of valve 52 is grounded and the control grid is also grounded through resistor R57.



   Resistor R57 is shunted by a capacitor C45, the time constant of this combination being such that the valve
52 is engaged, and the relay A actuated satisfactorily, by each interruption caused by the operation of the dial. The purpose of resistor R56 is analogous to that of resistor R55, namely to limit the flow of current through diode D8.



   Before indicating the purpose and describing the arrangement of the mechanically actuated switch and the switches in the circuit of a line detector, reference will be made to the selectors * Examination of fig. 3d shows that each selector is connected to the main or trunk lines
G and R of the multiplex, via the trunk or trunk crossings 410, 430 and 420, 440 which are analogous to the crossings 41, 43 and 42 and 44 of the line detectors. Further, the circulation system and the pulse introduction arrangements are also analogous to those of line detectors. There is, however, a difference in the release arrangements.



   The control gate potentiometer of the non-conductive valve '470, of the 460,470 trigger valves in a selector, is connected

 <Desc / Clms Page number 33>

 at one end to the anode of the other trigger valve 460 and, at the other end, to the tap of a second potentiometer R67 and R66.



  This second potentiometer is connected between a supply voltage Va which is negative with respect to V5, and the terrestrial phantom of the SG pair.



   When the potential of said terrestrial phantom is high by the call and busy signal, sent from the associated line selector, the point of ptise of the potentiometer R67 and R66 is raised substantially to the voltage V5, condition in which the control device trigger is able to function satisfactorily, but when said call and busy signal is removed, the voltage at the tap point of R67 and R66 is reduced to the point that the trigger device is unable to. function.



   The valve 54, fig. 3d to its control grid connected to the terrestrial phantom of the SG pair and its cathode to the supply voltage V3.



  The anode of the valve 54 is connected to a plug in the potentiometer R403. R440 and R450, controlling the introduction passage 400 as in conductor X connected to analog potentiometer R43, R44, R45 in the line detector circuit.



   In the release condition, the valve 54 is not conductive and the cathode of the valve or introduction passage is then positive to such an extent, with respect to the associated control grid, that introduction is prevented. When, however, the call-and-occupy signal is extended from the line detector, valve 54 becomes conductive and thus changes the polarization of the cathode of the valve or introducer passage so that introduction is then permitted.



   It is now clear that if a selector is prepared, upon receipt of a call and busy signal from the associated line detector, any positive direction path pulse of the correct amplitude appearing on the line main or common junction SM, will be introduced, and then put into circulation in said selector which, because of the main line or junction crossings, will be

 <Desc / Clms Page number 34>

 now connected to the subscriber to whom a synchronous path pulse has been reserved with that appearing on the main line or on the SM junction line and introduced into the selector at valve 400.



   The call and busy signal, present on the earth phantom of the SG pair, raises the gate bias of valve 420 of FIG. 3 to the operating point, as will be described, by causing this passage to emit a path pulse in the main or trunk line R, and this path pulse being synchronous with that of a particular subscriber, it will be sent and, by causing the release of relay S in its exchange, will determine its call.

   When the called subscriber answers, he causes transmission in the main line or G junction line connected to the selector in FIG. 31,7-the path pulsation reserved for it, which thus passed through the valve 410 of the suitable selector and, making the valve 510 non-conductive in a manner analogous to that described with reference to the line detector of the fig. 3c, determines an increase in the potential of the terrestrial phantom of the SR pair. This, as described, results in the release of the relay S of the calling subscriber, thus giving the answer signal,
When a calling subscriber is connected to a line selector and forms the number of the desired subscriber, the path pulse specific to the desired subscriber is caused to appear on the main or trunk line SM.

   Referring to fig. 3a, the valves 21 to 30 have a common anode resistor R68 and are connected by the resistor and the decoupling combination R69 and C50 to the supply voltage V1. The cathodes of the valves 21 to 30 are earthed and via the control grids are connected individually, / a resistor, to a supply voltage V4, negative with respect to the earth, to a sufficient extent, to cut the associated valves. The control gates are also connected by conductors M1 to M10 to the DS2 contact series of DS digit switches of the line detectors.

   The valves 21 to 30 have their stop grids directly connected to those of the multiplex modulating valves 1 to

 <Desc / Clms Page number 35>

 
10 respectively, and if the voltage of the control grid of the valve 21 is raised by the application of the earth potential to the contact of the corresponding series, in any of the digit switches, the valve 21 will emit then a negative sense pulse in the common anode circuit, the pulse of this pulse being synchronous with the pulses of the subscriber path pulse associated with valves 1 and 11.



   Similarly, by applying the earth potential to any one of the conductors IL to M10, the corresponding path pulses can be made to appear in the common anode circuit of the valves 21 to 30. pulses appearing in this anode circuit are first reversed in polarity by the amplifier
38 and then communicated, via the cathode load amplifier mount valve 39 to the main path or / junction SM. Thus, in order to connect a selector to a particular subscriber, it is necessary to apply the earth potential to the appropriate conductor M1 to M10 and at the same time to extend the call and occupancy signal. - tion from the associated line detector.



   The reference to fig. 3d shows that the eight-microsecond carrier frequency pulses emerging from the delay line output amplifier of the selectors engaged, in addition to providing the rectified pulsation necessary to maintain traffic, are communicated, via conductors PSO and decoupling resistors, to the main track or to the PS junction. As has been described, the pusatinsalnalogues circulating in the line detectors are also caused to appear on the main or junction line PS, and it is then clear that the transformer T6, served by the line main or PS junction, will transmit the versions, in carrier frequency, of the eight microseconds of the path pulses of all the subscribers engaged in the exchange.



   These pulses are rectified by diodes D2 and D3 in order to produce pulses of negative direction across resistor R9 and, as Rg is part of the total gate leakage resistance of valve 39 and because these pulsations occur in advance

 <Desc / Clms Page number 36>

 relevant path pulses and persist until after said path pulses have passed, they modify the gate bias of valve 39 in such a way that they prevent transmission in the main or SM junction line from any pulsation of journey already started. This is the means by which a called subscriber line is tested for commitment before being connected to a selector.



   Referring again to fig. 3c, when a subscriber is connected to a line detector, the normally non-conductive valve 53 is made conductive due to the connection between the control gate of this valve and the earth phantom of the SG pair.



  Relay E operates and at contacts E1 extends the potential, from earth, through conductor W, to the series of selector switches of the line marker.



  In addition to the above, the action of engaging a line detector produces the release of the normally actuated pulse acceptance relay, A. After a short delay, the slow protection relay B operates on contacts A1 and actuates the relay. common cutoff CO in the line marker by closing the contacts B1, thereby preventing the X-marking conductor of a line detector at contacts CO1 from taking the next call. As can be seen, the CO relay remains actuated for the time taken to make the connection between two subscribers.

   Operation of relay B also opens the arcing circuits DS3 at contacts B2, prepares the control circuit of the decipher switch at contacts B3 and prepares the holding circuit of relay MF at contacts B4. At the first interruption in the pulse train formed at the dial, relay A operates and completes the control circuit of the electronic control of the digit switch DS3 to contacts A1 via contacts b3, the rest contact of the arcing DS1 . the relay in series C to the DS control unit. The slow relay C. operates, closes contacts C3 and the DC relay operates. The switch moves at the end of the interruption, when relay A is released again.

   Subsequent interruptions to the train keep relay C in operation and move

 <Desc / Clms Page number 37>

 the switch via the arc DS1 and contact C4. At the end of the pulse train, relay C releases and shortly after relay 00 releases. In the meantime, that is to say the release time of the DC relay, the MF relay operates via the C1 and CC contacts and locks the MF2 and B4 contacts. The operation of the MF relay triggers the release of the CO relay at the MF contacts.



   Also, during the release of the DC relay, the ground potential is extended through contacts C2, CC2 and the arc DS2 to the appropriate conductor M1 to M10. Thus, depending on the number formed on the dial, the link is made with the desired subscriber In the event that this subscriber is engaged, no pulsation appears on the main line or on the SM junction during the release time of relay C and consequently no binding is made. When relay C releases, the control circuit is open at contacts C4, so that any further action on the dial is ineffective.
The release or cut of the link is under the control of the calling subscriber because when he opens the circuit of his line, the pulsation passed through the valve 41 of the line detector is reduced to zero amplitude.

   Relay A then operates and initiates the release of slow relay B.



   Valve 51 becomes conductive and the resulting final drop in voltage at the anode produces, via capacitor 046, a negative pulse at the control gate of valve 47 which, as has been described, prevents operation of the device. trigger and thus stop the circulation.



   In addition, the voltage drop at the anode of the valve 51 eliminates the call and occupancy condition of the earth phantom of the pair.
SG and thus stops the circulation in the selector, and also, by making the valve 53 non-conductive, releases the relay E and, thus, by eliminating the earth potential of the conductor W makes the detector loose, -length available for another call. At the same time, relay B releasing relay MF at contact B4 and, at contacts B2, completing the return circuit of the digit switch.


    

Claims (1)

CLAIMS. EMI38.1 ------------------------. 1. a multiplex communication system, comprising means for applying signals to a single circuit, a multiplicity of other circuits each of which is suitable for being connected, at spaced, desired time intervals, with said single circuit , passage means and means for intermittently actuating said passage means, so as to place said single circuit in operative connection with one of said other circuits at the desired time intervals. 2. A communication system which comprises means for applying time-spaced signals to a single trunk or trunk line, a multiplicity of intercom lines, a multiplicity of passage means each of which is associated with the connection. one of said intercommunication lines and arranged to command or control the connection of an intercommunication line with said main or junction line, and means for intermittently actuating a passage means so that a line of intercommunication is operatively connected to a trunk or trunk line for the transmission of signals from the trunk or trunk line to said intercommunication line at the time intervals at which the signals appear on the trunk or trunk line. 3. A communication system as claimed in claim 2, in which said passage means are arranged to be actuated by means which comprise a source of electrical pulses, conditioned to command or control a passage means constituted. by an electronic valve, 4. A multiplex communication system which comprises a main or junction line, an intercommunication line suitable for transmitting signals, to said main or junction line, a multiplicity of circuits, passage means associated with the EMI38.2 known as inter-communication line agenoés to command or control <Desc / Clms Page number 39> the connection of said intercommunication line with the main or junction line, and means for intermittently actuating said passage means, so that said intercom line is functionally connected to the main line or junction for the transmission of signals, from the intercommunication line to the main or junction line, at time intervals such that the trunk or trunk line is intermittently connected to a desired circuit. 5. A multiplex communication system which comprises a main or junction line, a multiplicity of circuits each of which is suitable for transmitting signals spaced in time to said main or junction line, a multiplicity of intercommunication lines , a multiplicity of passage means, each of which is associated with one of said intercommunication lines and arranged to command or control the connection of an intercom line with said main or junction line, and means to intermittently actuate any of said passage means, so that the associated intercommunication line is functionally connected with the main or junction line for the purpose of transmitting signals from the main or junction line to said intercommunication line associated with the time intervals at which signals appear on one of the said circuits. b. A multiplex communication system which comprises a multiplicity of circuits, all capable simultaneously of transmitting signaling pulses to a single main line or junction, a multiplicity of intercommunication lines, a multiplicity of means of passage, each of which is associated with one of said intercommunication lines and arranged to command or control the connection of an intercommunication line with said main or junction line, and means for intermittently actuating each of said passage means so that the associated intercommunication line is operatively connected to said main or junction line for the purpose of transmitting signals from the line. <Desc / Clms Page number 40> main or junction to said intercommunication line associated with the time intervals at which signal pulses appear on one of said circuits. 7. A multiplex communication system which comprises means for generating electrical pulses spaced apart in time, a multiplicity of calling circuits arranged to transmit audio-frequency signals, means for modulating said pulses by said signals audio frequency, so that the signals emanating from any of said calling circuits modulate pulses separated from each other by fixed, predetermined time intervals, a main line or common junction arranged to transmit modulated pulses emanating from all said call circuits, a multiplicity of intercommunication lines, a multiplicity of passage means each of which is associated with one of said intercommunication lines . and arranged to command or control the connection of an intercommunication line with said main or junction line, means for intermittently actuating each of said passage means so that the associated intercommunication line either functionally connected with the main or junction line for the purpose of transmitting signals from said main line or junction to said intercommunication line associated with the time intervals at which pulses, which correspond to one of the so-called call circuits, appear on the main or trunk line, and means for demulating the pulses received by an intercom line, whereby audio-frequency signals are transmitted on the intercom line . 8. A multiplex communication system as claimed in claim 5, or claim 6, which comprises another main or single trunk line, suitable for being connected with each intercom line, which is operably connected to the first line. main or junction mentioned, by other means of passage associated with said intercommunication line - a multiplicity of other circuits and means for actuating tion, <Desc / Clms Page number 41> intermittently said other means of passage in order to connect the main line or junction mentioned last to said intercom line at time intervals determined by means associated with one of said other circuits, 9. A multiplex communication system as claimed in claim 5, or claim 6 which comprises a multiplicity of called circuits, a main or common line connected to said called circuits, other passage means associated with each. said intercommunication lines and means for intermittently actuating said other passage means to connect said last-mentioned main or junction line with an intercommunication line at time intervals determined by means associated with one of said called circuits, so as to determine the appearance of modulated electric pulses on said main or trunk line with a view to transmission to one of said called circuits. 10. A multiplex communication system as claimed in claim 9, which comprises means for demodulating pulses received on any of the called circuits, whereby audio-frequency signals are transmitted on the called circuit. 11. A multiplex communication system in which audio-frequency signals appearing on any one of a multiplicity of speech circuits, all suitable for being connected simultaneously with a main or common trunk line, are caused to appear on the main or common trunk line as time-spaced pulses and are transmitted on any selected of a multiplicity of intercom lines which are connected to the trunk or trunk line by passage means actuated to connect the main or junction line, functionally, with said intercommunication line at the time intervals at which pulses corresponding to signals on a conversation circuit appear on said main line or junction. <Desc / Clms Page number 42> 12. A multiplex communication system as claimed in any one of claims 4 to 12, wherein the passage means are arranged to be actuated in response to signals on the circuits which are suitable for transmitting signals to the line. main or junction. . 13: A multiplex communication system as claimed in claim 12, which comprises a multiplicity of switching devices, each of which is associated with one of said intercommunication lines which are suitable for being connected with the network. main line or common junction line, each of said switching devices being able to actuate passage means for only that of the intercommunication lines with which the switching means is associated. 14. A multiplex communication system as claimed in claim 13, wherein each switching device is conditioned to be responsive to signals applied to the common trunk or trunk line by any of said circuits connected to the trunk. said main or trunk line, means being provided to prevent any switching device from being sensitive to signals originating on more than one of said circuits at a time. 15. A multiplex communication system as claimed in claim 14, comprising means arranged to, upon actuation of a switching device, transmit a signal over the immunization line to which said device. switching is associated when said device has responded to a signal on the main line or common trunk. 16. A multiplex communication system as claimed in claim 15, wherein said means are arranged to modify the permanent voltage state of the intercommunication line when said switching device has responded to a signal on the communication line. the trunk or trunk line, whereby another switching device associated with said intercommunication line is conditioned to be sensitive to appearing signals. <Desc / Clms Page number 43> sant on another main line or common junction, 17. A multiplex communication system as claimed in any one of claims 13 to 16, which. comprises one of other switching devices, multiplicity / each of which is associated with one of said intercom lines, each of said other switching devices being able to actuate passage means associated with the single line of communication. intercommunication with which said other switching device is associated. 18. A multiplex communication system as claimed in one of claims 13 to 16, which comprises means suitable for making it possible to make the switching device associated with a single intercom line sensitive at all times to signals. originating on one of the circuits which apply signals to the main line or common trunk. 19. A multiplex communication system as claimed in any one of claims 6 to 18 which comprises a multiplex device suitable for determining that signals on a number of circuits appear as pulses spaced apart in time. , and applying the pulse formed by pulses corresponding to signals on any circuit, to a common trunk or junction line, in such a manner that successive pulses corresponding to signals emanating from any circuit, are spaced apart by a time interval during which pulses corresponding to signals from the remaining circuits may appear on the main or common trunk line. 20. A multiplex communication system as claimed in any one of claims 13 to 19, wherein each switching device comprises a thermionic valve circuit capable of being conditioned by a line marker device to respond to the first. pulse of a pulse appearing on the main line or common junction. 21. A multiplex communication system as claimed in claim 20, wherein the switching device comprises means for rendering said thermionic valve circuit incapable. <Desc / Clms Page number 44> to respond to impulses which follow said first impulse, other than impulses of the same pulse, 22. A multiplex communication system as claimed in claim 21, wherein the switching device comprises a pulse generator adapted to generate an electrical pulse under the control of said thermionic valve circuit when said last-mentioned circuit responds to said first pulse, and means responsive to the pulse from said pulse generator, arranged to prevent a passing thermionic valve from allowing pulses of the pulse to which a switching device has responded, to be applied to any other switching device. 23. A multiplex communication system as claimed in claim 22, wherein the pulse generator comprises an electronic trigger circuit and which comprises means for re-establishing said trigger circuit after it has been operated. under the control of the thermionic valve circuit. 24. A multiplex communication system as claimed in any one of claims 20 to 23, wherein said thermionic valve circuit is adapted to control the passage means by which an intercom line is connected with a main line or common junction for the transmission of signals from the main line or junction to the intercom line. 25. A multiplex communication system as claimed in any one of claims 23 to 25, in which the trigger circuit is adapted to be re-established or readjusted by pulses originating from a source other than the means of switching, and wherein said thermionic valve circuit is arranged to be caused, by means responsive to the pulse produced by the pulse generator, to cause the pulse generator to produce another pulse. 26. A multiplex communication system as claimed in any one of claims 20 to 25, wherein the passage means for denying an intercom line to the <Desc / Clms Page number 45> main or junction line, comprising a pair of thermionic valves, one of which is suitable, under the control of said thermionic valve circuit, to allow the other valve to place the intercommunication line and the main line or junction in communication, 27. A multiplex communication system as claimed in any one of claims 23 to 26, wherein the pulse generator is arranged to cause a voltage to be applied to the thermionic valve circuit, in order to prevent said circuit from responding to any other pulse except a pulse controlled by the pulse generator. 28. A multiplex communication system as claimed in any one of claims 23 to 25, wherein a pulse from the trigger circuit is conditioned to control the application of an alternating current pulse to a delay device, for transmitting a delayed pulse to said thermionic valve circuit. 29. A multiplex communication system as claimed in claim 28, wherein a change in the continuous voltage of the intercom line produced when a switching device has been actuated is caused to condition another device. switching to respond to signals applied to it by a multiplex device arranged to generate pulses corresponding to the various pulses of said called circuits. 30. A multiplex communication system as claimed in claim 29 which comprises means for causing the application of selected pulses produced by the multiplex device to said other switching device, said means being responsive to pulses transmitted by an inter- communication line, 31. A multiplex communication system as claimed in any one of claims 19 to 28, wherein the multiplex device is arranged to generate current pulses. <Desc / Clms Page number 46> DC under the control of any one of a number of speech circuits, and comprising means for modulating said direct current pulses using signals from said speech circuits. 32. A multiplex communication system as claimed in claim 31, wherein the multiplex device comprises means for generating further pulses which coincide in time with the pulses corresponding to signals on a circuit of a multiplicity. circuits suitable for applying signals to the main or common junction line, and means responding to a predetermined rate of pulses produced on one circuit of said multiplicity of circuits to apply another. selected pulse to a switching device, whereby said switching device is arranged to actuate passing means permitting the transmission of signals from an intercommunication line to a desired circuit. 33. A multiplex communication system as claimed in claim 32, which comprises means for preventing the application of one of said other pulses to said switching device in the event that a desired circuit is already connected with a line. intercomniunication ,, 34. A multiplex communication system as claimed in any one of the preceding claims, in which the connection of any one of a multiplicity of circuits capable of carrying signals is effected by a main or trunk line common to an intercom line selected from a multiplicity of intercom lines by the operation of electronic switching devices having physically stationary members or parts. 35, A multiplex communication system constructed and arranged to operate substantially as described with reference to and as illustrated by Figures 3a 3b, 3c, 3d and 4 of the accompanying drawings.