CH286286A - Installation of telephone exchange. - Google Patents

Description

Installation of telephone exchange. The present invention relates to a telephone exchange installation.

According to the invention, the telephone exchange installation is characterized by distributing means allocating to each of a plurality of telephone communication paths a determined short time interval in a series of continuously repeating short time intervals, from to represent each communication channel by a series of pulses which repeat themselves with a determined frequency, by a plurality of Link circuits comprising one of them an electronic line searcher to search for a calling communication channel and to establishing a circuit leading from the calling communication channel to a called communication channel, and by a blocking circuit responding to the establishment of said circuit to prevent other Link circuits from selecting a communication channel Taken by the established.

It is noted from the above that the invention relates to an at least partially electronic telephone exchange installation using electronic line searchers and that, thanks to the invention, the double test of line searchers is avoided. lines. The installation which is the subject of the invention could also be entirely electronic, as will appear from the description which follows. The accompanying drawings represent; by way of example, one embodiment of the telephone exchange installation forming the object of the invention.

fig. 1 is a diagram, showing the general layout of the circuits.

Figs. 2 and 3 are views of the dispenser tube employed in the installation.

Figs. 4 to 8 are electrical diagrams of the circuits of the installation.

fig. 4 represents the common equipment.

fig. 5 shows the pulse shaping apparatus.

fig. 6 is the first line finder suit.

fig. 7 is the recorder circuit. fig. 8 is the selection circuit. The assembly of figs. 4 to 8 must be done in accordance with the diagram of fig. 1 - it is also indicated by fig. 9.

fig. 10 provides a number of curves to explain the operation of various parts of the plant, and fig. 11 represents, in section, a type of delay line which may be suitable for making the circuits of FIG. B.

In the example shown, one can subdivide the circuits into three parts, as shown in FIG. 1. First, all RTI ID="0001.0267" WI="4" HE="4" LX="1583" LY="2189"> subscriber lines, twenty in number for example, numbered from 1 to 20. Each of these lines has the usual terminal equipment 21 at the subscriber's. The second Part is the common equipment designated under the number 22. The third Part is constituted by a group of Link circuits, one of which is necessary to establish each communication.

Each of the Link circuits may, in turn, be subdivided into several circuits represented by special reference numbers in FIG. 1, namely: the line searcher circuit 23, the dial pulse shaping circuit 24, the dial pulse recording circuit 25 and the line selection circuit 26. Circuits 23 to 26 are numbered 27 to 38.

For the sake of simplicity of description, an will deal only with communication in one direction, that of the other direction hui being similar.

As seen in FIG. 1, all subscriber lines terminate at the common equipment, 22, to which a scanning function is assigned, preferably by means of a cathode ray tube, the beam of which rotates regularly.

When one of the subscriber lines is marked with a potential Lin which indicates that this subscriber is calling, the common equipment 22 applies to the wires 27 and 28 signals which can act on all the Link circuits in parallel, 1' we will szippose, for example, that Ast the first of these circuits which will come into service. The line searcher 23 of this Link circuit will pick up the calling line and then transfer the signals, via wire 33, to the dialing Pulse shaping circuit 24.

As the rotational signal rises, circuit 24 generates rotational pulses which are counted and stored in recording circuit 25, which then serves to control selector circuit 26, which carries a time shift line. or delay line or other devices suitable for this purpose.

Speech signals pass through common equipment 22; then arrive, by wire 28, at the line searcher circuit 23, from where they leave, by the wire 33, to cross the selection circuit 26 from which they come out by the first wire 36 to return to the common equipment 22 of where, finally, they are applied to the outgoing line.

The part of fig. 1, which has been enclosed by a broken line and which contains the line searcher circuit 23, the numbering pulse shaping circuit 24, the recorder circuit 25 and the selection circuit 26, can be considered as a unit playing the role of a Liaison circuit between the two subscribers in communication.

For some system applications, a peLit sync frequency is sent from common equipment 22, over wire 29, to selector circuit 26 as well as to line searcher circuit 23.

The three wires 27, 28, 29 (coming out of the common equipment) as well as the two Pils 36 and 37 (entering the common equipment) are used on all the Link circuits, as seen in FIG. . 1.

The rotary distribution role to be played by common equipment 22 may be provided by a beam tube, as seen in FIGS. 2 and 3.

The distributor tube 39 carries an aethode 40, the usual grid 41, the electrode 42 serving both the role of beam concentrator and anode, the horizontal deflection plates 43 and the deflection plates Vertical 44. Phase current from a suitable source is supplied to the Deflector Plates by conductors 45, 46, 47 and 48, so as to drive the beam in a circular rotational motion.

The electronic target of the tube 39 is divided into twenty specific targets 49 to 68, each of which is coupled with one of the subscriber lines 1 to 20. The targets 49 to 68 are constituted by secondary emission electrodes. daire which are associated with a common anode 69, so as to form dynodes having a common output. A mask or screen 70 can be placed in front of the beams, with suitable openings so that the beam only strikes each dynode when it is well aligned with it and the dynode is thus sheltered from the action. of the secondary emission of the other dynodes.

Tube 39 exits via anode 69 and wire 71, then via the circuits which will be described later and which, via Pils 27 and 28, lead to the line finder circuit.

Currents from selection circuit 26 are applied through wire 36 to grid 41 of tube 39 and serve to modulate the electron beam in accordance with the energy of the selected signal.

Thus, referring to FIG. 1, it is seen that the energy leaving the tube 39 by the conductor 71 can be reapplied after a suitable delay of time (produced in the block 26, as will be explained later), by the conductor 36, at the grid 41 of tube 39 to provide the communication path between the two lines to be connected.

The common equipment 22 is represented by FIG. 4. It has been assumed for purposes of explanation that the frequency of 10,000 cycles per second has been chosen to obtain the scanning of the dynodes. This frequency is high enough for all the vocal frequencies useful for the transmission of speech to be reproduced with fidelity.

The Base frequency is derived from a Moor 72 oscillator at 200 kilocycles and preferably stabilized by a crystal. It is preferable to use a high frequency master oscillator, because stability is then much easier to achieve. Moreover, in certain variants of the installation, the frequency of 200 kilocycles can be used directly for certain commands. The sine frequency generated in the master oscillator 72 is reduced to the base frequency of 10 kilocycles by means of a frequency divider 73.

The currents from the frequency divider 73, suitably phase shifted by 90 by a phase shifter 74, are applied respectively to the sets of vertical and horizontal deflection plates 43 and 44 of the distributor tube 39. The beam of this tube will rotate at 10,000 revolutions per second, and each of dynodes 49 to 68 will be scanned once every ten thousandths of a second.

Subscriber lines 1 to 20 are connected to the dynodes. 0n has represented In fig. 4 the connections of lines 1, 5 and 20 on the dynodes 49, 53 and 68. Line 5 is also represented with the subscriber unit 21, of the conventional type. Microphone 75 is connected in series with dial dialer 76 and hook switch 77. Receiver 78 is <RTI ID="0003.0273" WI="6" HE="4" LX="1685" LY="680 "> permanently bridged On the line in a stream of simplification, which makes it possible not to represent the ringing devices assuming, for example, that the called subscriber can be alerted by a special tone sent in 1st receiver 78.

At rest, the switch hook contact 77 is normally open. When the subscriber makes a call, he picks up and the contact 77 closes, thus completing a circuit which borrows the loop of the subscriber's set, the low-pass filter 79 and the battery 80 whose negative pole is thus applied to the dynode 53.

Normally, all of the dynodes 49 to 68 are at the same potential as the common anode 69 and no current passes through them. The negative potential applied to the dynode 53 will result in a secondary emission by this dynode when the beam from the tube 39 sweeps over it, i.e. there will be at this moment a negative pulse on the lead wire. exit 71.

The pulses are preferably modulated by the signals only at a depth of 25 to 501/o, so that there always remains sufficient amplitude to supply the energy which is necessary for the establishment and maintaining connections, regardless of modulating signals.

Negative pulses resulting from the potential on dynode 53 are supplied to the grid of inverter tube 81. The anode circuit of tube 81 is coupled to the grid of peaker tube 82 which, by removing the threshold of these brings them to a pre-determined level so that only unmodified Parts are transmitted. Thus, at the outlet of tube 82, the conversation currents can be effectively modulated at 100 1/o.

The pulses thus treated are then applied to tube 83 and, through the outlet of this tube, to all the connecting circuits via wire 28.

On the inverter tube 81, a second output is made on the cathode side, and the pulses coming out of this side are applied to a clipped tube 84 which cuts off the high s of the pulses and brings them to a constant level, so as to eliminate modulation effects. The anode circuit 84 is coupled to the grid of a cathodyne tube 85, which in turn applies the pulses 86, through the resistor 87, to the wire 27, by which they will reach the grid of the tube 88 which is the electronic gate tube. of the line searcher 23 (fig. 6). (Electronic gate is the equivalent of electronic gate.) Since all the line searchers of all the link circuits are in parallel, the pulses 86 are applied at the same time to all the all tubes 88.

Pulse 86 issuing from tube 85 passes through resistor 87, and the pulse issuing from resistor 87 will be more specifically called pulse 89. The pulses applied to the grids of the tubes 88 are more precisely, pulses 89. Under the present conditions, that is to say if all of the tubes 88 of the line searchers are not nearly as strong as pulse 86, but if the assumed conditions change, pulse 89 will become much weaker than pulse 86, as will be explained later.

In the absence of any signal on the cathode of a tube 88, the pulse 89 applied to the grid of this tube is insufficient to pass Plate current, because the Bias voltage of the grid of this tube is much lower than the cutoff value. In the line finder 23 (see figs. 1 and 6) there is an oscillator 90 which provides a frequency somewhat lower than Gelle which is supplied by the frequency divider 73 of fig. 4. This oscillator can, for example, operate at a frequency 0.02% lower than the frequency of the divider. Energy from oscillator 90 is applied to clipper amplifier 91 which serves to produce square pulses 90a. , These pulses 90a RTI ID="0004.0274" WI="7" HE="4" LX="1431"LY="676"> are differentiated by passing through a network consisting of a capacitor 92 and a resistor 93 and they take the form 94 in which they are applied to the control grid of clipping tube 95. to the cathodyne tube 97, from which issue pulses 98 applied in a first turn to the cathode of the tube 88 and tending to make this cathode more negative, that is to say to make the tube 88 more conductive.

However, tube 88 does not function so long as the arrival of pulses 98 does not coincide in time with the pulses 89 which are applied to its grid by wire 27. The grid of tube 88 is , in fact, negatively biased quite strongly by Battery 99 and it requires the simultaneous combined action of the two pulses 89 and 98 to operate this tube.

Since oscillator 90 is outputting at a frequency somewhat different from that of divider 73, there will come a time when pulses 98 will begin to coincide in time with pulses 89 arriving from the calling line. At this time, the bias effect of tube 88 will be overcome and pulses 100 will occur in lead 32 of the tube. These pulses 100 pass through capacitor 101 and act at 102 on a sharp peak amplifier and on a Phase corrector circuit, the combined role of which is to trigger oscillator 90 to pulses 89, so that From this moment this oscillator will be in synchronism with the divider 73 and the pulses 98 will coincide irregularly in time with the pulses 89 sent by the calling line.

As soon as the oscillator 90 is on and Pulses 100 pass regularly over the wire 32, these Pulses 100 are also applied to the rectifier 103 and, behind it through the integrator network 104, to the control grid of tube 105 which, on operation, will increase the positive voltage on the grid-geran of tube 95, thus increasing the amplitude of the Mit of this tube, i.e. the amplitude of the Pulses 96 and 98.

The value of the resistor 87 (see fig. 4) and the characteristics of the tube 88 are such that the total variation of its grid voltage with respect to the cathode gel cannot exceed a small amplitude fixed in advance, whatever whatever the amplitudes of the Pulses 98 and 86 applied respectively to the cathode and to the grid (through the resistor 87). However, the change in the Polarization of tube 95 a increases the amplitude of Pulses 98. Thus, if the sum of the amplitudes of Pulses 98 and 89 is to remain approximately decrease.

The purpose of this decrease in the amplitude of the Pulses 89 is to prevent the operation of the tubes 88 in the other line seekers associated in parallel with the one which has just operated. This will be better explained later when examining fig. 10.

The decrease in the amplitude of the pulse 89 does not decrease, however, the output of the link circuit tube 88 engages, since the total voltage difference between grid and cathode remains unchanged. Thus, Pulses 100 remain approximately constant in amplitude.

Pulses 100 exiting tube 88 via wire 32 are also applied, through coupling circuit 106, to an electronic Gate control tube 107 which serves to control the suppression Bias of the Input Gate tube 108. This tube 108 is normally biased at its stop grid, so that pulses reaching it through wire <B>28</B> (from common equipment circuit tube 83) cannot 1st cross. However, when the tube 107 is operated by the Seizure of a calling line, as just described, the polarization of the stop gate is removed and it becomes conductive during the time intervals which correspond to the opening of the the way of the calling line. Consequently, <B>IQ</B> tube 108 will pass, during these intervals, on its output wire 33, pulses which may be combined pulses of dial and conversation 109, towards the device. impidsion conformation 24 (Figs. 1 and 5) and to the line selection equipment called 26 (Figs. 1 and 8). However, RTI ID="0005.0268" WI="7" HE="4" LX="1120" LY="997"> to the line selection equipment of FIG. 8, the energy will not pass yet, until the selection of the called line has been made. This will be explained later.

The line searcher 23 having now fulfilled its role, the Pulses 109 coming from the calling line During the time intervals when this line is open will be applied, via the line 33, to the integrating network 110 of the impulse shaper (fig. 5). This network may be preceded by a circuit which spreads the pulses and is analogous to a peak voltmeter.

Integrator network 110 functions as a low pass filter which passes dial pulses, but cuts out higher frequency communication signals. At the output of the network 110, the pulses are amplified by the tube 111, then pass through the transformer 112 and are applied to the control grid of the collector tube 113 and also to the control grid of another tube 114.

The tube 113 cuts off the Dial Pulses which now arrive in order to transform them into rectangular Pulses 115, which, in turn, are differentiated at the exit of the tube 113 by Zn network 116 and applied to the control grid of the electron gate tube. - only numbering 11.7.

The tube 117 is polarized so as to suppress all the negative part of the pulse, differentiated which arrives at it (the part which corresponds to the front flank of the rectangular pulse 115) and not to let pass the positive part of the differentiated pulse, corresponding to the rear edge of the rectangular pulse 115. Normally, the tube 11.7 is located near the cocapure due to the voltage drop in the resistor 118 which is connected to its grid. -screen. Resistor 118 is in the Plate circuit of a normally conducting tube 119 forming part of a tilting circuit (Eceles-Jordan) and working in conjunction with tube 114. The time constants of this circuit are such that the flanks before the First Dial Pulse Operates tube 114 and therefore shuts off tube 119. Low pass filter 120 which is then in the gate circuit of tube 119 maintains this Si tuation until the last dialing pulse is passed and, shortly thereafter, the base circuit returns to its normal state and the Electronic Gate 117 becomes insensitive again.

Thanks to this special blocking circuit, transient effects before and after dialing pulses cannot affect the recorder.

Tube 117, when in operation, delivers over wire 35 to the recorder circuits (circuits 25 of Figs. 1 and 7). 0n see nur l.a fig. 7 that the pulses sent to the recorder by wire 35 pass through the resistors 121 and 122 before being applied to the grids of the tubes of the first stage of the recorder.

The circuits of the recorder comprise a series of tubes of which only a few, 123, 124, 125 and 126, are represented in detail in fig. 7. These tubes are coupled in pairs to form binary counters operating on the principle of relaxation circuits. The dem-. Groups 123-124 and 125-126 are shown in detail, bloss 127, 128 and 129 are similar relaxation pillows. There must be a sufficient number of relaxation circuits to be able to count all the call numbers which will be transmitted. In the hypothesis of the twenty-line desk where Fon West places the five relaxation circuits of fig. 7 should suffice e.

At rest, the right tubes of each group, such as 124 and 126, are conductive and polarize the left tubes 123 and 125 i the cut-off voltage of the Plate current.

The voltages created in the recording circuits are applied to the output pins 130 to 139 and through them to the row select circuits of FIG. 8, to polarize therein, as will be explained later, the electronic gate tubes of the delay circuits up to their plate current cocapture voltage or else the electronic gate tubes without delay for the to make conductors.

The pulses arriving at the recorder through wire 35 are applied to the First binary group comprising tubes 123 and 1.24, which is assumed to be at rest, i.e. with tube 124 conducting and tab 123 polarized. at the plate current cocapture voltage. In this case, an output voltage is applied to the output wire <B>130</B> which, in accordance with the selection circuit of FIG. 8, will maintain the Electronic Gate tube without delay in operation; a voltage also exists on wire 131, by which the electronic delay gate tube will be blocked, as will be explained below.

The First pulse arriving through wire 35 passes through resistor 121 and acts on the grid of tube 124 which is biased at the plate current cocapture voltage, but tube 123 is rendered conductive and the tube operates. 123 will apply to the Pils 130 and 131 a control voltage which blocks in the selection circuit the electronic gate tube without delay 142 and will open the electronic gate tube with delay 153.

The output of tube 124 is connected by wire 140 to the second set of binary counter tubes of the recorder, comprising tubes <B>125</B> and 126. Will be transferred from tube 126 to tube 125 or vice versa each time group 123-124 returns to its normal condition (conductive tube 124). It is therefore clear that the group 125-126 changes its condition at the arrival of one pulse out of two, while the group 123-124 changes its condition at the arrival of each pulse.

The third group 127 of two tubes is analogously controlled by wire 141, and this group 127 will change its condition each time the group 125-126 returns to its normal condition, i.e. each time 126 becomes conductor. Group 127 therefore changes its condition once for every two operations of group 125-126.

The fourth group 128 of two tubes will similarly change its condition each time the third group 127 returns to its normal condition, and the fifth group 129 is again controlled in the same way by the output of the fourth group 1.28.

Referring now more particularly to FIG. 8, the operation of the recorder tubes to control the delay time will be fully explained. To fully understand the mechanism of these operations, it should be known that the dials of the tuners 76 of the subscriber stations (FIG. 4) are dials with twenty holes and that the number of pulses emitted by these dials, when an les arme On the number of one of the twenty lines, corresponds to the difference between the number number of the calling line and the number of the line called.

It is now necessary to produce the displacement in time of the energy of the communication, displacement which must correspond to the difference in the instants of scanning of the calling line and of the called line in the cathode ray tube 39 The impulses transmitted by the requesting subscriber will act in the recorder of FIG. 7, as already has Bit.

Each of the five Good recorder circuits has respective access to five electronic gates without delay 142 to 146, as well as to five electronic gates with delay 148 to 152, which have different delays: 5 microseconds for the 148, 10 microseconds for the 149 , 20 microseconds for the 150, 40 microseconds for the 151 and 80 microseconds for the 152. These figures will be sufficient in the hypothesis of a twenty-line office.

Each of the electronic delay gates has a delay line followed by an electron gate tube. Two of these electron carrier tubes only, 153 and 154, have been drawn in fig. 8 (the others should be drawn in a similar way).

At rest, before the arrival of the Pulse clipping, the polarizations are such that the Electronic Gates without delay 142 to 146 are all open, so that the Pulse clipping 1.09 coming from the line finder of FIG. 6 by wire 33 passes without delay through the five tubes of circuits 142 to 146 in Series and from there by wire 155 will be applied to the output tube 156. Assuming for the moment that the tube 156 is not blocked, its Plate circuit will then deliver a Pulse 157 which, through the wire 36, would reach the control electrode of the tube 39 (fig. 4) and fall back on the calling line.

When the first tube of the recorder operates, for the first time, the control potential is transferred from wire <B>130</B> to wire 131, making tube 153 conductive, while tube 142 is polarized ä, the cut-off voltage. Thus, the First Registered Pulse introduces a delay of 5 microseconds and the energy arriving via wire 33 now passes through delay path 148, then through the four non-delay paths 143-146, in Series.

The second pulse transfers the control potential of wire 131 to wire 130, which makes circuit tube 142 conductive again, while tube 153 of delay path 148 will turn off. At the same time, as explained above, the second recorder circuit operates and transfers the control potential from wire 132 to wire 133, which blocks circuit tube 143 and opens circuit tube 154 from second circuit. delay line 149, which introduces a 10 microsecond delay between line 33 and line 155, with power passing through 142, 149, 144, 145 and 146.

The third pulse will not affect the second recorder circuit, but will act again on the first recorder circuit, so that the passage of energy between wire 33 and wire 155 will take place through 148, 149 , 144, 145 and 146 with a total delay of 5 microseconds in 148 plus 10 microseconds in 149, or 15 microseconds.

The fourth pulse will reset the First as well as the second recorder switch, but will activate the third recorder switch 127 on the other hand, which will create a delay of 20 microseconds in 150 for the Passage of 1'. energy from wire 33 to wire 155.

The fifth pulse will add 5 microseconds to the previous 20 microseconds. The next pulse will replace the 5 microseconds (taken at 148) by 10 microseconds (taken at 149) and the Energy Passage Delay Will then be 30 microseconds. The next pulse will add 5 microseconds (taken at 148) to the previous 30 microseconds. The next pulse will render delay elements 148, 149 and 150 ineffective and replace them with element 151 worth 40 microseconds to itself, and so on... each new pulse introducing 5 microseconds of additional delays by suitable additions or mutations of the delay elements 148 to 152. to the nineteen others in the Office of Twenty Lines.

If the Bureau were to have more than twenty lines, it is obvious that additional groups of two tubes and additional groups comprising electronic delay gate tubes and electronic delay gate tubes could be added to 1 recorder and selection eireuit.

After the nine-rotation, the energy in pro-; origin of the calling subscriber's set Will be transmitted, as explained, as pulses from the common equipment circuit, through the wire 33 of the link circuit, the delay elements and the wire 155, to the grid. tube 156 (Fig. 8). Output pulse 157 from tube 156 passes through wire 36 to tube control electrode 39 (Fig. 4). The vocal modulation of the pulses 157 arriving through the wire 36 will cause variations in the electron current of the tube 39 each time the electron beam is aligned with the particular target of the called line. These variations will be transmitted on the called line through the low pass filter 79 of this line and. will reach the 1st RTI ID="0008.0267" WI="15" HE="4" LX="1328" LY="1104"> telephone receiver 78.

To make the call, a special frequency can be transmitted and a call receiver tuned to this frequency can be tuned at the subscriber's premises or, if the receiver 78 is in permanent bypass, it may suffice to whistle at the subscriber's. calling subscriber or making any other conventional noise that the calling receiver 78 could reproduce with sufficient intensity to alert that subscriber.

It was assumed in the foregoing that tube 156 was conductive. But this tube is normally biased at the plate current cut-off voltage, so that dialing pulses arriving through circuit 23 cannot affect other lines. The cut-off voltage of tube 156 is controlled by a circuit comprising tubes 158 and 159 (FIG. 5). Tube 158 is normally conductive and the grid of tube 156 is then biased by wire 38 to the plate current cut-off voltage.

The two tubes 158 and 159 are driven by tube 119 (Fig. 5) in the following way: Tube 119 biases to the plate current cut-off voltage at the beginning of the series of dialing pulses. At this time, tube 119 sends capacitor 160, and subsequently the grid of tube 158, an ineffective positive pulse. But as soon as the emission of the Series of dialing pulses is completed, the tube 119 becomes conductive again and sends a negative pulse which cuts off the operation of the tube 158 and makes the tube 159 conductive and, consequently, the 1st tube 156. This then allows power applied to this tube by wire 155 to pass and exit through wire 36 to the called subscriber's line.

To prevent the called line from being seized by line finders on other trunks when the called subscriber picks up, a portion of pulse 157 is branched from wire 36 by wire 37 and passed through a resistor <B> 161 </B> (see fig. 4) specific to each link circuit to an occupation pulse shaper 162 whose output is applied to the grid of an electronic occupation gate 163 which forms part of the common equipment. This limits the maximum possible value of the positive pulse 89 coming out of tube 85 when the called subscriber goes off-hook. The thus reduced value of the positive pulse 89 will be insufficient to actuate the electron gate tube 88 of a line finder.

When the conversation is over and the calling subscriber hangs up, the recorder circuits of FIG. 7 as well as the output control tubes 158 and 159 of FIG. 5 must be returned to the resting position. This is effected by tubes 164, 165 and 166 of FIG. 7. When the line finder 23 had been engaged, the tube 105 of FIG. 6 had received a cutoff voltage and as a result the gate voltage applied by wire 31 to tube 164 had decreased. This operated the flip-flop circuit consisting of tubes 164 and 165 and the conductivity condition was transferred from tube 164 to tube 165. A negative pulse was then sent through wire 167 and capacitor 168 to tube 166. , but as this tube 166 was polarized at the cut-off voltage of the current Plate, this negative Pulse had no effect. On the contrary, when the calling subscriber hangs up and the line finder hangs up, the tube 105 of FIG. 6 becomes conductive again, the Bias applied by wire 31 to tube 164 rises and the flip-flop circuit 164-165 returns to its normal state, which sends a positive Pulse to tube 166, which passes through this tube and decreases. the potential to resistor 169, which is enough to reset the recorder circuits as well as the two tubes 158 and 159.

* To avoid harmful reactions between the div ers. circuitry of the recorder as well as with the tubes 158 and 159, it is necessary that the resistance 169 be quite low. It is necessary, therefore, that the first tube 164 can debit a fairly intense current and, if necessary, an can first constitute it by a group of several tubes in parallel.

To better explain the operation of the System, we will now follow a call from line 1 for line 5. When the calling subscriber picks up his receiver, the hook switch 77 (FIG. a negative potential is applied to dynode 49. When the electronic beam from tube 39 falls on this dynode, a secondary emission will occur and result in a pulse in the common vanode 69. This pulse then passes through the inverter tube 81, 1 the clipper amplifier 84, the cathodyne tube 85, the resistor 87 and, through the wire 27, arrives at the electron gate tube 88 of the line searcher (fig. 6). Tube 88 then passes Pulses 100 which serve to lock oscillator 90 onto the calling line. The Pulses 96 drawn from this oscillator (conformed to become Pulses 98) are then synchronous with the Pulses 89 (those which the EI 27 brings to the tube 88). Because of this incidence, only Pulses 89 which correspond to the particular path of the calling line can pass through the tube 88 and generate Pulses 100. The other Pulses 89 which would come from other calling or called lines are removed from the pipe. operation of tube 88. Pulses 100 exiting tube 88 serve to drive tube 107, which causes electronic input gate tube 108 to conduct at appropriate times. Pulses 109 exiting tube 108 through wire 33 are spiked with the result of a selection among all Pulses sent by anode as 69.

The calling subscriber then forms the digit 5, which will produce a train of four reductions in the Bias of the dynode 49. The Pulses leaving the distributor tube 39 by the wire dynode 49, four successive amplitude reductions. These pulses are transmitted by wire 71 to the Plate circuit of inverter tube 81, clipper tube 82, tube 83, and wire 28, through which they reach the line finder input tube 108 control grid. (fig. 6). Due to the effect of the clipper tube 82; the four amplitude reductions in the Series will appear as four complete interruptions in that Series. The four line openings by the dialer dial of the requesting subscriber will therefore be reproduced at the outlet of the tube 108 and transmitted by the wire 33 in the form of pulses 109. The pulses 109 pass through the integrating network 110 (FIG. 5) whereby the opening interrupts are transformed into dial signals, which are amplified in tube 111 and pass through transformer 112 and receiver tube 113 (giving them the rectangular shape 115). . Their shape is further modified by the network 116, then they reach the electron carrier tube of the n-Lunerotation imptsion 117, cross this tube 117 and, by the wire 35, go. act on the recorder's request.

At the same time, the dialing signals pass through the integrating network 120 (FIG. 5) to activate the mechanism for unlocking the tube 117. This mechanism carries the tubes 114 and 119, the first conductor and the second polarized. at the cut-off voltage when the signals arrive. The action of this mechanism increases the positive bias on the tube 117, so that the impulses 115 can easily pass through this tube.

The Pulses 115 which have passed through the tube 117 will control, in the particular case envisaged, the operation of the first three circuits of the recorder, the third of these circuits remaining in the engaged position, while the first and the second, after being engaged momentarily, will have returned to idle. This introduces into the circuit (see Figs. 7 and 8) the delay element 150 has a value of 20 microseconds and corresponds to the time lag of the positions of the electronic beam of the tube 39 between the targets. 49 and 53 respectively associated with subscriber lines 1 and 5.

Simultaneously, increasing the voltage of tube plate 119 causes a positive pulse in capacitor 1.60 to circuit 158 and 159, but this pulse exerts no action, leaving tube 158 conducting and maintaining the tube 156 locked During the dialing time.

As soon as the dialing is completed, the positive potential is removed from the grid of tube 114 and the circuit constituted by the two tubes 114-119 returns to its rest position, in which the tube 119 is conductive. This has the effect of lowering the voltage on the screen grid of tube 117 which will no longer allow any signal to pass to the recorder fig. 7. At the same time, the decrease in the voltage of tube plate 119 causes a negative pulse in capacitor 160 towards the circuit constituted by the two tubes 158 and 159, which goes into the working position, ear tube 159 - becomes a driver. This one, by the wire 38, unblocks the electronic gate tube 156 (output electronic gate tube of the reader of fig. 8<B>)</B>.

The talk signal pulses 109, arriving via wire 33, are applied, through the selector delay circuit and wire 155, to tube 156, which passes them over wire 36 to control. der grid 35 of tube 39 and modulate in amplitude the electron beam of this tube. This modulation matches the talk signals sent by the line each time the electronic beam is directed onto the target 53 of the line 5. The low pass filter 79 which is on the line 5 restores the talk currents which go to receiver 78 of subscriber 5.

When the conversation is over and the calling subscriber hangs up his receiver, the line loop is opened and the negative potential ceases to be applied to electrode 49. Henceforth, when the electron beam passes over target 49, it does not A pulse will be sent on wire 71 and the link with the line finder will be interrupted. At the same time, the linefinder tube 105 will no longer deliver and the oscillator 90 hookup control will no longer function. The line hunter is therefore freed up and ready to serve other calls.

The existing potential ä. This moment on tube 105 is applied, through wire 31, to tubes 164 and 165 of the recorder release circuit. This group of tubes returns to its resting condition, in which tube 164 is conductive, causing a positive pulse in capacitor 168 to tube 166. This operates and applies a return to rest potential to the common resistor 169 through which the ciruits of the recorder pass in turn, which return in turn to rest, so that the electronic doors without delay are henceforth alone open in the selector. The control circuit for the Electronic Gates tubes 158-159 of FIG. 5 is also put back to rest (t.ube 158 conductor).

Thus, the entire engaged link circuit has been reset.

We have already pointed out that the Ale plays through the tube 105 of the line finder circuit (FIG. 6) to reduce in amplitude the pulses coming from the seized line (line 1 in the preceding example), in order to that other line finders cannot enter this line twice. It is clear from the foregoing description that when two subscribers or a greater number of subscribers are communicating at the same time, there passes a number of cooked from the common equipment of fig. 4. All of these pulses exit through wire 28 (output wire of tube 83) and are applied in parallel across the circuits of RTI ID="0011.0277" WI="11" HE="3" LX="1572" LY ="484"> desktop binding. When a link circuit has been engaged, however, it is necessary that the pulses intended for this circuit become ineffective for the other link circuits. The curves in fig. 10 will help to better understand the mechanism of this operation.

Pulses from anode 69 of tube 39 are applied to the grid of tube 81 which has two separate outputs: a plate output and a cathode output. On the Plate output, the pulses (which vary in amplitude according to the instantaneous signal) are represented by the curve 10A, they are processed by the first tube 82 which cuts the first threshold at level 170 and only lets through the amplitude modulated part 171 which goes to tube 83. It has already been pointed out that the energy should not be modulated, preferably more than 25%, so that the variations caused by the modulation are of interest only to a small Part of the total energy of the pulses. The pulses which pass through the Plate output of tube 81 are those which will be used for the conversation between the two subscribers, but they are not the ones whose study interests us the most at this moment.

The pulses leaving through the cathode of tube 81 are of major interest. They are emitted by the tube 8.4 and sent to the cathode tube 85, from which they emerge in a series of pulses of equal amplitude 86, represented by the curve 10B of FIG. 10. Pulses 86 pass through resistor 87, ä. the output of which, having become pulses 89, they are applied in parallel to all the grids of the electronic gate tubes 88 of the line searchers (fig. 6).

In the line finder, the oscillator 90 produces a wave 172 represented by the curve 10C of FIG. 10 and whose Period is slightly longer than the time interval which separates two consecutive pulses 89. The alternating wave 172 is doubly erupted at both levels 173 and 174 (FIG. 10) and then shaped by suitable reseatixes to produce rectangular pulses 96 at the leading edge which coincides with the point at which the increasing ground 172 passed through the value zero. These pulses (which are substantially wider than the pulses 89) pass through a cathodyne tube 97, at the output of which they become pulses 98 which are applied to the cathode of the electron gate tube 88 of the linefinder.

As the frequencies of pulses 89 and 98 are slightly different, the phase of pulses 89 will be constantly varying with respect to that of pulses 98, and there will come a time when these two phases will coincide, as seen on curve 10D. of fig. 10. When this occurs, the electronic Gate tube 88 will be activated and the Pulses can exit through the Plate circuit of this tube 88, pass through the sharp maximum amplifier 102 and hook on the oscillator 90. The correction of Phase introduced by the amplifier 102 is such that the sinusoidal ground 172 generated by the oscillator 90 crosses the zaro while increasing a little before the end of the arrival of each pulse 89. The pulses 98 will then be generated in relation to Constant phase with Pulses 89, as 1e represents - curve -10E of fg. 10 in his first wave of Bauche. - Once these pulses have been synchronized, the electronic gate tube 105 closes, which increases the value of the voltage -de. Biasing the screen grid -<B>of</B> tube 95, so that the amplitude of the Pulses 98 will increase. The Pulses 98 thus change from the search amplitude to the amplitude which they will maintain as long as the line searcher remains engaged. The curve 10E of FIG. <B>10-</B> represents the result of this increase in amplitude - (see second wave from the left) of pulse 98 which is to reduce the effective width of pulse 89.

Thus, from this moment, the pulses 89 which are applied in parallel to all the grids of the electronic gate tubes 88 of all the line searchers will become very weak, as represented by the third wave from de la Bauehe On curve 10E. 16th if, in one of these searchers, there was a coincidence between the pulse 89 thus reduced and a search pulse 98, the electronic door tube 88 of this line searcher would not be unblocked, as may be the case. See by examination of fig. 10E (fourth wave from La Bauche).

When the subscriber called has answered, the loop of his line closes and the negative potential of battery 80 is applied to dynode 53 (remember that this is subscriber No. 5). . If these special precautions were not taken, this would lead to the triggering of another line searcher who would enter this subscriber in duplicate. To avoid this eventuality, the System has included the Busy Configurator 162 and the Electronic Busy Door Tobe 163 which operate in the following manner: tube 156 of Fig. 8) lets pass the conv ersation pulses <B>157</B> which leave this tube by the first wire 36 to excite the grid 35 of the distributor tube 39. Part of the energy Pulse 157 is branched off by wire 37 and resistor 161 (Fig. 4) to the ocenpation pulse shaper 162 which forms part of the common eqtlipeinent. There they are amplified, flattened and reconstituted as strong rectangular Impulses of constant amplitude. (It is necessary, for this purpose, that the clipping level of the tube 82 should be such that the vocal modulation never reduces the Pulses 171 below a certain minimum). These Pulses are applied to the gate of Electronic Occupancy Gate Tube 163 and render this. highly conductive tube. The operation of the tube 163 has the effect of limiting the amplitude of the pulses 89 and of reducing this amplitude to a value insufficient to cause the called line to be seized by a second line searcher. However, this limit must remain high enough to keep a line finder who has already seized the called line in Seizure, otherwise the need to dial to an already occupied line would result in the communication of this line being cut off.

0n will now examine the fug. 11, where a delay line is shown for cases where a slight delay is necessary. For very short delay times of 5, 10 or 20 microseconds (case of electronic devices 148, 149 and 150), artificial lines of the conventional form can be used. For higher delay times, it. It is preferable to use a line consisting of a tubular receptacle 175 filled with mercury 176 and whose length will be equal to. Z X D, Y being the speed of sound in the liquid and D the proposed delay. At the entrance and at the exit of the line there is a crystal, for example a quartz crystal <B>177</B> or 181, mounted in a ring 178 or 182 with an electrode 179 or 183 to which the wire 180 carrying the electrical signal, or an electrode 183 connected to the starting wire 184.

From the foregoing, it is obvious that the installation is provided with all the necessary means to ensure the establishment of a communication between a requesting subscriber and a requested subscriber without the two lines concerned being able to block other line finders than that which connects them.

CLAIM Telephone exchange installation, characterized by distribution means granting each of a plurality of telephone communication channels a short time interval determined in a series of short time intervals which is repeated continuously, so as to represent each channel of communication. communication by a series of pulses which are repeated with a determined frequency, by a plurality of link circuits each comprising an electronic line searcher to search for a calling communication channel and to establish a circuit containing the calling communication channel to a called communication channel, and by a blocking circuit Responding to the establishment RTI ID="0013.0276" WI="9" HE="3" LX="1561" LY=" 541"> of said circuit to prevent other link circuits from selecting a communication channel taken by the established circuit.

Claims (7)

SUB-CLAIMS 1. Installation according to claim, characterized in that said blocking device comprises First means (88; 90,<B>1.05)</B> responding to call signals from the calling channel to prevent other link circuits from selecting said calling channel, and second means (162,<B>163)</B> for preventing the link circuits from selecting the called channel when said called channel answers. 2. Installation according to claim, characterized in that the line finder of each link circuit comprises selection means normally testing said paths in succession, an electronic circuit (88) responding to the coincidence between the test carried out by said selecting means and the call pulses caused by the calling channel to stop the progress of said selecting means, whereby the selection of said calling channel results, and a control device responding to the selection of said channel to reduce the efficiency of pulsing the selected channel for other line searchers. 3. Installation according to sub-claim 2, characterized in that said selection means comprise an oscillator (90) whose frequency is slightly different from the frequency of repetition of the channel pulses, and in that said communication device - Order comprises means (88, 102) for putting said oscillator in synchronisine avee the pulses of said channel. - 4. Installation according to sub-claim 1, characterized in that said first means are arranged so as to increase the sensitivity of the Link circuit selected for said call signals. 5. Installation according to sub-claim 3, characterized in that said oscillator is arranged to produce Blocking Pulses, and in that said electronic circuit (88) is a mixer in which said Blocking Pulses and said Unblocking Pulses are mixed, said mixer outputting Pulses when the Calling Pulses and the Unblocking Pulses coincide, further characterized by a Synchronizing device responsive to the Pulses outputted by said mixer to synchronize the Pulses unblocking with said determined repetition frequency whereby said calling channel is continuously selected, and in that said control device comprises means (103, 105) responsive to said continuous selection of said calling path to increase the amplitude of said Release Pulses for the purpose of increasing the sensitivity of the selected Link circuit to Call Pulses whose effectiveness is reduced by other link circuits. 6. Installation according to sub-claim 1, characterized in that said second means are arranged in such a way as to apply blocking impidions to said Link circuits, the position of which in time and the repetition frequency correspond to the channel selected. , so as to prevent the selection of the route called by a line searcher. 7. Installation according to sub-claims 4, 5 and 6.