US2952742A - Dial impulse register - Google Patents

Description

p 1960 ZENICHI KIYASU l-n-AL 2,952,742

DIAL IMPULSE REGISTER ll Sheets-Sheet 1 Filed May 31, 1956 '/Z PHA UM S A SAAM& A L omANoo w T UN A mm m Vm A WNS IU EOUHR ZK mA I UU SB UOW. SNT U Sept. 13, 1960 ZENlCHl KlYASU ETAL 2,952,742

DIAL IMPULSE. REGISTER Filed May 31, 1956 11 Sheets-Sheet 2 ,'g 2Q x x J g Z flgzflll 1 11 I[[' I llllHll llHlllH 111mm H HH IIIHIII Hmm I Ill Hlllllll mmm m: H mm! mum w m lllHllll lHlHlH lllllHll mum m m IH'IIYI Att s.

Sept. 13, 1960 ZENlCHl KIYASU ETAL 2,952,742

DIAL IMPULSE REGISTER Filed May 31, 1956 ll Sheets-Sheet 3 INVENTORS ZENICHI KIYASU. KOSHIRO HANAWA. SUSUMU KATSUNUMA. NOBUICHI IKENO & TAKEHARU FUKUOKA VIWfKJ WM Sept. 13, 1960 ZENlCHl KIYASU EIAL 2,952,742

DIAL IMPULSE REGISTER Filed May 31, 1956 ll Sheets-sheet 4 'g- 4 Id d DIAL cu/r/rmr 4 INVENTORS ZENICHI KIYASU. KOSHIRO HANAWA, SUSUMU KATSUNUMA. NOBUICHI JKENO 8. TAKEHARU FUKUOKA Sept. 13, 1960 ZENICHI KIYASU ETAL I 2,952,742

DIAL IMPULSE REGISTER Filed lay 31, 1956 11 Sheets-Sheet 5 NOBUICHI NO &,

TAKEHARU FUKUOKA BY 7142M, \(PM At y Sept. 13, 1960 Filed May 31, 1956 ZENICHI KIYASU ETAL 2,952,742 DIAL IMPULSE REGISTER ll Sheets-Sheet 6 ]I III I II III a Q Q m7 0; M9

INVEN TORS ZENICHI KIYASU,

KOSHIRO HANA WA, SUSUMU KATSUNUMA' NOBUICHI IKENO8- TAKEHARU FUKUOKA Sept. 13, 1960 ZENICHI KIYASU ETAL DIAL IMPULSE REGISTER ll Sheets-Sheet 7 Filed May 31, 1956 mvznrons ZENICHI KIYASU. KosmRo HANAWA. susuuu KATSUNUMA. uoawcm -|KNO a, TAKEHARU FUKUOKA Y v fwafiu w Sept. 13, 1960 ZENlCHI KIYASU ETAL DIAL IMPULSE REGISTER ll Sheets-Sheet 8 Filed May 31. 1956 A MM 0 K wm u Fmfi A U I R A H E M T Y B Sept. 13, 1960 ZENICHI KIYASU ETAL "I 2,952,742

DIAL IMPULSE REGISTER Filed May 51, 1956 11 Sheets-Sheet 9 ZENICHI K! YAS U. KOSHIR O HANAWA,

SUSUMU KATSUNUMA, NOBUlC HI IKENO & TAKEHARU FUKUOKA BY WM, AttJS.

Sept. 13, 1960 ZENICHI KIYASU ET AL DIAL IMPULSE REGISTER Filed May 31. 1956 ll Sheets-Sheet 10 fig-7 INVENTOR'S ZENICHI KIYASU. KOSHIRO HANAWA, SUSUMU KATSUNUMA. NOBUICHI IK'ENO8. TAKEHARU FUKUOKA WM, 2:}. 1 $0M United States Patent Office DIAL IMPULSE REGISTER Zenichi Kiyasu, Koshiro 'Hanawa, Susnmu Katsunuma, Nobuichi Ikeno, and Takeharu Fukuoka, Tokyo, Japan, assignors to Nippon Telegraph & Telephone Public Corporation, Tokyo, Japan, a corporation of Japan Filed May 31, 1956, Ser. No. 588,407

7 Claims. (Cl. 179-18) This invention relates to a dial impulse register, and more particularly to a dial impulse register in which parametrically excited resonators are used as unit operating elements thereof.

It has hitherto been the practice to register the subscribers dial impulses by relay devices or mechanical devices, in cross-bar or other telephone exchange systems. However, the cross-bar switching system requires many relays and therefore is expensive. Other mechanical systems are defective in that the sliding parts thereof are easily worn, and the flexible parts thereof are easily damaged, and the adjustment thereof is difi'icult. Furthermore, in view of the present-day speeding-up of the common control parts thereof, the above systems are considered to be already at their maximum operating speed.

One object of the present invention is to provide a dial impulse register, in which the operating elements thereof make no movement, whereby such operating elements will not be worn or damaged. The dial impulse register according to the invention therefore has a much longer life than the known dial impulse registers.

Another object of the present invention is to provide a dial impulse register which has a high operating speed, and. in which a dial having a higher dial-speed or a larger dial speed variation than the known dials, may be used.

In order to accomplish the above objects, there are used parametrically excited resonators, formed by nonlinear reactance elements, as unit operating elements of the dial impulse register, and to count and store the dial impulses by electric circuits which are formed by a combination of such resonators. Such resonators are composed of electric elements which make no movement, and have no practical limitation of their life, in which property they differ from the electronic elements such as vacuum tubes and discharge tubes, in which the thermionic emission is decreased during normal life.

The accompanyingdrawings show for the purpose of exemplification, without limiting the invention or claims thereto, certain practical embodiments illustrating the principles of the invention wherein:

Figs. la-d and 2a-g are circuit diagrams and symbols for explaining the operating principle of parametrically excited resonators.

Figs. 3a and b are the circuits for converting the dial current into high frequency current having two kinds of phase-states.

Fig. 4 is a circuit for generating timing signals, necessary for measuring make and break periods of dial current, together with a block diagram of the converting circuit of Figs. 3a and b.

Fig. 5 is a circuit diagram for detecting the interdigital pause.

Fig. 6 is a circuit for counting impulses.

Fig. 7 is a circuit for counting a number of decimal digits.

Fig. 8 is a circuit for storing the number of impulses Patented Sept. 13, 1960 counted, namely registering the oflice number and the called subscribers number.

Fig. 9 is a connecting line between the circuits of Figs. 7 and 8.

Fig. 10 is a diagram showing how the circuits of Figs. 4 to 9 can be combined together.

Figs. 11 and 12 are other embodiments of the counting circuits.

We shall explain hereunder the construction and the performance of a parametrically excited resonator, the unit operating element of the present invention. In the resonance circuit of Fig. In, L is a magnetic core, such as a ferrite core, having coils thereon, C is a capacitor, and R is a resistor. A high frequency current having a small amplitude and frequency f (for example, 1 me.) is applied to the terminal 3 of the above resonance circuit, and then a curernt having frequency 2 and a comparatively large amplitude (hereinafter called the exciting current) is applied to the terminals 1 and 2. An oscillating current having frequency 7, half of the exciting current, is produced in the resonance circuit. This phenomenon is caused by the periodic variation at 2 of the inductance in the resonance circuit due to the excitation by the current of frequency 2 and the phenomenon can be explained mathematically by the solution of Hills equation or Mathieus equation, if the current which is produced in the resonator is small (see E. T. Whittaker and G. N. Watson: A Course of Modern Analysis, pp; 404-428, Cambridge Univ., 1935. In actual practice, the magnetic cores have considerable saturation characteristics, and therefore, the oscillation produced is not so simple as Mathieus solution. However, the essential oscillation is quite the same. Due to the saturation characteristics of magnetic cores, the above oscillating current produced in the resonance circuit reaches a steady state having a constant amplitude. The phase of the oscillating current, produced in the resonance circuit, is limited to either of the two phases, difierent from each other by 7r radians, as shown in Fig. 1b. Which of the two phases the oscilalting current takes depends upon the initial condition, namely, the phase of the high frequency current applied from the terminal 3 to the resonance circuit. However, the oscilaltion phase is always either one of the two phases, different from each other by 71' radians. We shall hereinafter refer to these two phases as 0 phase and 1r phase respectively as shown in the drawings, or simply as 0 and 1, in logical representation.

In Fig. 1a, the windings between the terminals 1 and 2 are balanced so that the voltage induced in the resonance circuit by the exciting current will be cancelled, and the exciting current having frequency 2 applied at the terminals 1 and. 2 does not appear in the resonance circuit. A gain of scores of decibels can be obtained between the oscillating current having frequency 1 produced in the resonance. circuit andthe current applied to the terminal 3. This is of advantage in making circuit connections in cascade. or in branches, without using amplifiers composed of vacuum tubes or transistors.

Fig. 1c shows the case in which there are provided three input terminals, 3, 4 and 5, instead of the single terminal 3 in Fig. 1a. In this case, it will be clearly understood that the phase of the oscillating current produced in the resonance circuit, after the impression of exciting current, is determined by the phase of the superposed current of the three currents, which are applied to the terminals3, 4-and 5. Therefore, the phase of the oscillating current in the resonance circuit is as shown in the- Table I, depending upon the phases (ll-phase or 1r-phase) of the three high frequency currents having equal amplitude and appliedto the terminals 3, 4 and 5.

Table II represents such and rr-PhflSBS replaced by logical signs, 0 and 1 respectively. By fixing the phase to be applied to the terminal 3 as 0-phase or rr-phase, a logical product circuit (hereinafter referred to as AND and shown as 6) and a logical sum circuit (hereinafter referred to as OR and shown as 69), of which the input is applied from the terminals 4 and 5, are respectively formed, and the output thereof can be taken out fi'om the terminal 6.

TABLE II f p R in Figs. 1a and c is a coupling resistor of the input or output of the resonance circuit. However, instead of such resistor, an element with any impedance or a transformer may be used for the coupling element. Fig. 1d

shows an example of the transformer coupling. A single turn is sufficient as a primary winding of the transformer T, when the frequency is about 1 me. This makes the connections much simpler, and also makes the phase reversal (logically, it means NOT) simpler, it being necessary only to reverse the direction of the turn wound on the magnetic core. The resistors R in Fig. 1d are so selected that the transmission loss between two resonance circuits in adjoining stages will be about 30-40 db and that the producing and damping period of oscillation will be suitable. The unit elements formed by magnetic cores, capacitors, and resistors, shown in Figs. 1a, 0 and d are hereinafter referred to as parametrically excited resonators or parametrons, and the symbols in Fig. 2 shows such parametrons and their logical functions. Namely, Fig. 2a shows a single parametron; Figs. 2b and 0 show respectively parametrons having an AN function and OR function; and Figs. 20?, e and show the connection of a plurality of parametrons. In the drawings, the cross on the line connecting the two parametrons shows the phase reversal NOT. I, H and HI appearing in Figs. 2d, 2 and 1 show the excitation periods of the corresponding parametrons, and the three excitationperiods are made to overlap each other slightly in relation to time, as shown in Fig. 2g. By such excitations, the logical operation progresses following the order, III, II-III, and IIII. In the drawings, it is assumed that the logical operation normally progresses to the right. The back coupling to the left is specially marked by an arrow.

InFig. 3a, coils, L and U wound on the non-linear core, D and coils L and L wound on the non-linear core D are connected in a modified lattice form, L and L' having the same number of turns, L and L' having the same number of turns, and the number of turns of the former two being one more than that of the latter two. Therefore, the inductance of coils, L and U is larger than that of L and L' and the phase of current at the output terminals 3 and 4 is in phase with that of the current at the input terminals 1 and 2. However, when a direct current flows in another coil L wound on the core D the inductance of coils L and U decreases and becomes smaller than that of coils L and U due to the non-linear saturation characteristics of the core D and the phase of the current at output terminals 3 and 4 becomes in opposite to that of the current at'the input terminals 1 and 2. By connecting the two parametrons 7 and 9 with the above circuit, the parame- 4 tron 9 oscillates at a 0-phase state and a rr-phase state in response to the break and make of the dial contact of the subscribers telephone if the signal of parametron 7 is fixed as O-phase.

In Fig. 312, P and P show two parametrons which are coupled by resistors R1, R and a transformer T. In this case, P and P are coupled in opposite phase relation by resistor R transformer T, capacitor C, rectifier G and the earth, on the one hand, and are coupled in the same phase relationship by resistor R transformer T and ground, on the other hand. When the direct current does not flow in the rectifier G, the resistance of G is high, and the signal transmitted from P to P through the circuit, R T-CGground, is weaker than the signal transmitted from P to P through the circuit, R T. Therefore, P and P oscillate with the same phase. When a contact is made at the subscribers phone, a negative direct current of about 1 m.a. flows through the circuit, battery(48 v.)-resistor R -telephone contact-resistor R G. Then the resistance of G is lowered by scores of ohms. As the result thereof, the signal from P to P through R --TCG,-ground becomes more intense than the signal through R -T, and P is oscillated in with a phase opposite to the phase of P In this way, parametron P can be made to oscillate at O-phase state and vr-phase state, in response to the break and make of the current caused by a subscribers dialing.

In Fig. 4, the oscillation phase of the parametron elements 11, 12 and 13 is cyclically reversed from 0 to 1r or vice versa, during every period of one cycle of excitation III-III. Assuming that 11 oscillates at 0- phase during the exciting period III, 12 oscillates at 0- phase during the following exciting period 1, due to the in-phase coupling from 11. During the following exciting period II, 13 oscillates at vr-phase, due to the reverse coupling from 12. During the following exciting period III, 11 oscillates at 1r-phase, the same as 16. During the following exciting period I, 12 oscillates at rr-phase. During the following exciting period II, 13 oscillates at O-phase. It is clear that, in this Way, the oscillation phase respectively of 11, 12 and 13 is reversed at every cycle ,of excitation I]I-III. The signal alternately produced is applied from the element 11 to the elements 14 and 17. The circuits which follow, 1415.16; 18-19 20; 424344 respectively form loops. The oscillation phase of 14 is not changed when the signal from 11 is 0. However, when the signal from 11 is changed into 1, because 14 is an OR element, the phase of 14 is changed into 1. This signal cycles to 15 and 16, and this state is sustained even when the phase of 11 is returned to 0. In this case, the phase of -17 stays as 0, because 17 is an AND element, and unless the phases of 11 and 16 become simultaneously 1, the phase of 17 does not become 1. Then, when the phase of 11 is again changed to 1, the AND element 17 becomes 1 for the first time. And when the above signal 1 of the element 17 is applied to 18 and 21, the AND element 15 returns to 0, due to the NOT signal 0 applied to 15, and 16 and 14 are also changed into 0. Namely, whenever .the loop 14--1516 receives the phase 1 from 11, the

phase of this loop becomes alternately 0 and 1. In other words, at every two applications of the signal from 11, the signal 1 is produced at 17. The operation of the next loop, 1819-20 is quite similar. On application of the signal 1 at 17, the phase of the loop, 181920, is alternately changed into 0 and 1. At the element 21, the signal 1 is produced at every four occurrences of the signal 1 at 11. The operations of the following loops,

22-23-24, 26-2728, 4243-44 are also similar. Namely, the signal 1 is produced at the element 25 at every eight occurrences of the signal 1 at 11, at the element 29 at every sixteen occurrences of the signal 1 at 11 at the element 45 at every two hundred fifty-six occurrences of the signal 1 at 11. The element 5 11 produces the signal and 1 alternately, and therefore, assuming that the duration of exciting period 1, 11 and III is respectively 0.025 milli second '(ms.), the signal 1 is produced at 11 at every 0.05 milli second, and the signal 1 is produced a 45 at every 12.8 milli seconds, the duration of such signal being 0.025 milli second.

The value 12.8 ms. has been selected from the permissible range of the speed of dialing.

The signal output of the element 45 is applied to the elements 101, 106 and 111 in Fig. 5, through the lead 0, as a timing signal. The timing circuit of Fig. 4 can be used in common with other registers. The signal coverted by the circuit of Fig. 3, represented by box 8 and parametrons '7 and 9 in Pig. 4, is transmitted to the elements 101 and 106 through lead shown in the block diagram in the upper part of Fig. 4.

In Fig. 5, the converted dial impulse signals, which are transmitted from the element 9 to the'elements 101 and 106, are sampled by the timing signal which appears from the element 45 at every 12.8 ms. When the converted dial signal is -0 (corresponding to a break), the signal 1 appears at the element 101 at-the application of :the timing'signal with phase 1 thereto. When the converted dial signal is 1 (corresponding to a make), the signal 1 appears at the element 106 at the application of the timing signal with phase 1 thereto. When the .signal of the element 101 is -l, and if the signal of the loop, '1-03-104 105, was 1 during the period prior to the period at which the element 101 has a phase 1, the AND element 102 becomes 1, and this signal 1 is applied to the element .105 and causes the element 105 to become 0 which in turn causes loop '103104105 to become 0. When the dial signal has a phase 1 during the next timing period, the element 106 assumes a phase 1 and the phase 1 appears at the element 107 which changes the phase of 103104-105 into 1. Thus the loop 103104105 (the operation of which has been represented as that of 103) quantizes the make and break of the subscribers dial current and indicates the same as a signal in the form of a phase. Such quantizing accomplished by the above circuit is for avoiding miscounting, caused by the distortion of the duration of the make current and the break current which is due to the inductance and capacitance of the subscribers line and also by thesirnulated impulse which is due to chattering of contacts occurring in dial mechanisms. Since-theperiod of the timing signal is 12.8ms. in the device of this invention,a-correct counting can be made, without being influenced by a-simulated impulse or by an irregular break having a duration less than 12.8 ms. The elements 102 and 107 produce-signal 1 every 0025 ms., which is one timing period, only at the time when the signal of 103104-1-05 is changed from 1 to 0 or vice versa. The number of signals of phase 1, produced at the element 102, is equal to the number of impulses of each digit dialed by the subscriber. These signals are transmitted to the counting circuit of Fig. 6 .from .102. On the other hand, the signal 1 appearing at 107, which is produced when the dial contact is changed from make to break together with the signal appearing at 102, are utilized for the time scaling circuit, which will be explained below and is for detecting the interdigital pause by measuring the duration of the make and the -.duration of the break of dial current. The circuit formed by 108-128 is such a time scaling circuit. The elements 108, 109 and 110 are respectively connected to the elements 112, 116 and 120. For an easy understanding of the drawing, this connection is shown by R R and R Since the element 108 is an AND element and has a NOT input, 108 assumes a 0 phase when one of the elements 102 and 107 has a signal with phase 1, and assumes a phase 1 at other times. Therefore, the signal of 108 is 0 during the period .of 0.025 ms., every time the dial contact is changed from break to make, or vice versa. The signal 109 and 110 is exact- 'ly the same as that of 108, with a slight time :lag. 109 and 110 are provided for the purpose of producing a similar signal but with a time difference in the excitation period, in order to control the elements 112, 116 and 117, which have time difierences in their excitation periods. When the signal from 108, 109 and 110 is 0, ithesi'gnal of 112, 116 and 120 also becomes 0,'due to the fact that the signal from 111, and 119 is 0. (This is because, 111, 115 and 119 are AND circuits, and therefore, assume a 0 phase signal, except when the timing signal is applied to 111.) Then the phase of the loops, ill2 11311'4, 116--117118, 120121122, becomes 0, irrespective of what phase these loops :had during the previous stage. When the signal from 108, 109 and 110 becomes 1, 112, 116 and 120 act in the same way as an OR circuit. It will be understood that this is the same as the timing circuit shown in Fig. 4. The time scaling circuit 111-12S inFig. 5 is a binary counting circuit and counts the timing signals ever 128 ms. 'It forms a three stage binary counting circuit, and therefore, the signal of the element 123 becomes 1 during the period of 0.025 ms., only when the timing signals, 23:8 in-number, are applied thereto. Thus a signal with phase 1 appears at 123 after a lapse of 102.4 ms. (12.8 ms. 8) from a starting condition of =0 phase of the time scaling circuit. When this signal appears, the loop 124-1-25 126 takes it and sustains the signal 1, and the output from element is 'fed back to 111. Irrespective of the presence or absence of a timing signal, the phase of the signal of 111 is changed to 0, and stops the progress of operation of the time scaling circuit. If the signal of 102 or 107 becomes 1, prior to the lapse of 102.4 ms. after the signal 'of the time scaling circuit returns to 0, such signal 1 from 102 or 107 makes the signal of 108, 109 and 110 0, and returns the time scaling circuit 111-123 to 0 phase, irrespective of what phase condition the said time scaling circuit was taking. Atthe next timing signal, the counting is re-started automatically. In this case, the signal 1 does not appear at 123. This means that the duration of make or break of the dialcontacts is less than 102.4;ms. The above 102.4'ms. is a time duration which we selected in View 'of the speed of the present day telephone set, and therefore, the above time duration can be chosen at will by an engineer. The signal 1 appearing 'at 123 is applied to 127 and 128. Since 127 and 128 are both AND elements, the signal 1 appears at 127 for a period of 0.025 ms., 103 is producing 'a 1 signal. The sigial .1 from 103 indicates the state of make of a dial contact, and the signal 1 from 123 indicates the state continued for more than 102.3 ms., and therefore, such state can show an interdigital pause. The signal 1 appearing-at 127 is applied to the elements 154 in Fig. 6 and 161 :in-Fig. 7. When the signal from 103 is 0, namely, when 102.3 ms. has elapsed after the break of the dial current by the subscriber, the signal 1 appears at 128. This shows that the make and break cycle of dialing has stopped. By a timing scaling circuit as above explained, the "durations of make and break of dial contact can .be correctly counted. In the above example, the durations *of'mak'e and break of dial contact can be chosen between 12.8 ms. and 102.4 ms. This means that the dial-speed and the ratio of duration of make and break can vary widely,

and therefore, no precise mechanism is necessary for the phone dials. This makes the cost of production much less and makes it unnecessary to change dials due to their speed variation, the said speed variation constituting the greater part of the troubles occurring in the present-day telephone apparatus.

The circuit shown in Fig. 6 is similar to the binary counting circuit in Fig. 5. Compared with the circuit of Fig. 5, that of Fig. 6 is formed in the same way as far as the elements 131-446 are concerned. The only dif- 7 When the phase of the signal of two of the three loops, 136-131-138, 140141142 and 144--145-146, is 1, the phase of the signal of 147 becomes 1, and the element 135 is converted substantially from an AND element into an OR element.

Due to this when the next signal is applied to the counting circuit as input, the same state as 'if two input signals were received is introduced. The element 148 produces a signal with phase 1 only when none or one of the four loops, 132133134, 136137-138, 148141-142, and 144-145-146, has a signal of phase 1. The result obtained by counting the dial impulses in accordance with the above is shown in the following Table III:

TABLE III Number of impulses Loop 132-133- Loop 140-141- Loop 136137 138 Loop 144-145- Element 148 The above table shows that two loops (or elements) have a phase 1 and the other three loops (or elements) have a phase at every number of impulses (two out of five). Thus the decimal digits 1-l0 are each represented by one of the above binary signals, two out of five. Such a binary signal will, as is well known, be useful for detecting troubles during the subsequent operations. It will be understood that the counting of dial impulses by the two out of five form of signal is much easier than by the hitherto used relays, electronic tubes, etc. The elements, 154, 155, 156, 157, 158 and 159, change the phase of each loop to 0, to make each loop ready for the counting of the following digit, after the circuit of Fig. 6 has transmitted the counted digits to the circuit of Fig. 8.

Fig. 7 shows the case of a seven digit system (oflice number 3 digits and subscriber number 4 digits), and is similar to the above counting circuit. The number of digits dialed is represented as a binary signal by three loops, 162--163164, 166-167168 and 170171 172. The signal appearing at each loop, after dialing of By the logical product (so called AND) of three such loops, the elements 182-188 will take the following signal: Directly after the dialing of the hundreds digit of the oflice number, 182 assumes phase 1 for a short time, (0.025 ms.) and after the dialing of the tens digit of the oflice number, 183 assumes phase 1. Upon dialing the thousands, hundreds, tens and units digits of the subscribers number, 184, 185, 186, 187 and 188 respectively assume the phase 1 for a short time. These signals 'drive the parametrons 201207 respectively through the vacuum tubes V V in the storing circuit in Fig. 8.

' The elements, 179, 180 and 181, are for restoring the counting circuit when such elements receive the signal from 128 and the signal of the marker (from the terminal u), on the giving-up of subscribers dialing or on the completion of the desired connection with the called subscriber. V

Thirty-five elements, M M12, M in Fig. 8 are minute ferrite cores (diameter, about 2 mm.) having magnetic hysteresis. Two lead wires which are perpendicular to each other pass through each of the above cores. The parametrons appearing at the left side of Fig. 8, 149, 150, 151, 152 and 153, are respectively coupled with M 71 12 '12 M1a' 7s 14 74 311d M15 "M75, and the parametrons at the center of Fig. 8, 201, 202, 207, are respectively coupled with M -M M12M25, M17-M75. 1118 parametrons, have a frequency of oscillation, which is half of the oscillation frequency of the parametrons, 149-153. Parametrons 201207 are excited by the current which is obtained by amplifying respectively the outputs g g g; of elements 182, 183, 188 in Fig. 7 by means of the left half of vacuum tubes, V V V For such excitation, the oscillating currents from parametrons 182188 are applied to the grids, of vacuum tubes only when the signals g from 182 188 have a phase 1, and are not applied when the signal g from 182 188 have a 0 phase. This is accomplished by superposing currents of other parametrons having a 1r phase on the oscillating current g from parametrons 182, 183 188. Therefore, parametrons 201, 202, 207 are excited only when the signals from 182, 183, 188 have a phase 1, and are not excited at any other time. Directly after the dialing of the hundreds of the ofiice number by the subscriber, 201 is excited by the exciting current having a frequency 1. At this moment, a current having frequency f/2 flows in the secondary resonance circuit of 201. This current passes through the cores, M M and magnetizessuch cores with frequency f/ 2. The elements, 149153 oscillate at O-phase or 1r-phase with frequency f. Therefore, when the current of parametrons 149-153 and 201 is chosen properly, cores of M M are unsymmetrically magnetized, by reason of the superposition of a current with frequency f/2. from 207 on a current of frequency f and with 0- phase or 1r-phase from 149153, as shown in Fig. 8a. Thus, either one of the two states of residual magnetisation which may be present in a ferro-magnetic substance such as a ferrite core is placed in the core. This residual magnetism remains, even after the removal of the magnetisation currents, due to the oscillation current of 201. Which of the two kinds of residual magnetism remains depends upon whether the signal fi'om the parametrons 149, 153 has a phase of 0 or 1. Similarly, when 202 is excited (directly after the dialing of tens of the oflice number), the signal from 149153 is sustained in cores M M as residual magnetism. The same operation is repeated thereafter. Namely, the units of the ofiice number, and the thousands, hundreds, tens and units of the subscribers number are stored respectively in al- 35, 41 45 51* 55, MST-M65 and M71 M The operation all carried out in the form of two out of five code. In this way, all the digits are 'stored in M M- The stored signal information (ofiice number and subscribers telephone number) can be read in the following way: The signal g g g, is applied to the right half of each of vacuum tubes V V through the terminals, S S of Fig. 9. By means of such a current having frequency f, amplified by the vacuum tubes, V V the parametrons 201 207, are again excited, and the current having frequency f/2 is produced in each of such parametrons. We shall explain the case where the current having frequency f/2 is produced at 201. The produced current passes through the magnetic cores M M M Since the magnetic cores have non-linear properties, a voltage having frequency f, the second harmonic of f/2, is produced in reading can be made when the parametrons, .202, 203,

205, 206 and 207 are successively excited. Therefore, :parametrons '196200 are successively excited during the succeeding exciting .period HI, and oscillate in phase with the harmonics produced in each core. Thus, the-stored information can be read. The information of each core remains even attertlrezabovezreading. However, the dem'agnetisation thereof is :not necessary, because such remaining information is amalgamated with the succeeding information which is written at the next dialing.

In the example of-Figss6 and 8ythe counting and storing of digits dialed are carried out separately. The counting is done by the circuit of Fig. 6, and the storing is accomplished by using one core for each bit of information as shown in Fig. 8. The storing of the telephone number of the subscriber by using small magnetic cores, is stable and inexpensive.

It must be understood that the above explanation of the circuits of a dial impulse register is not to be limited to the examples shown. Other systems of counting, such as a conventional binary decimal code, decimal code and bi-quinary code, (all based upon a binary code) can be used in a manner similar to the above-mentioned two out of five binary code.

Fig. 11 shows the counting circuit in which the conventional binary code is used. The connection of parametrons is similar to that shown in Fig. 5, the only difierence being that a single parametron 301 replaces the AN element 111 in Fig. 5. Therefore, the circuit of Fig. 11 is always ready for counting of digits by dial impulses. The operation of this circuit is also quite similar to that of Fig. 5. It will be clear that the relation between the dial impulses and this circuit can be represented as in the Table V.

TABLE V Number oi Loop Loop Loop Loop impulses 302-303-304 306-307-308 310-311-312 314-315-316 No 0 0 0 1 1 0 O 0 2 0 1 0 0 3 1 1 0 0 4 0 0 1 0 l5 1 0 1 0 6 0 1 1 0 7 1 1 1 0 8 0 0 0 1 9 1 0 0 l 10=0 0 1 0 1 Fig. 12 shows the counting circuit in which the biquinary code is used. The circuit has six output terminals, I, 2, 3, 4, 5n and V of which u1 n5n assume phase 1, when a number of impulses 1 and 6, 2 and 7, 5 and 10 are respectively applied to the input of Fig. 12 and assume phase 0, at other times. At the terminal V," a signal of phase 1 appears when 6 or more than 6 impulses are applied, and a signal of phase 0 appears when 5 or less than 5 impulses are applied. To R, which is connected to the parametron 528, a signal of phase 1 is normally applied, and a 0 phase signal is applied only when the circuit is restored for the next operation. Once the phase of the signal of 528 is made 0, the phase of the signal of all the elements, 501-527, is returned to 0. Thereafter, 502, 505, 525, operate as if these were OR elements. In this state, when a signal is applied to the upper input terminal, the said signal is applied to 507, 501, 508, 511, 527. All the elements assume a 0 phase, except the AND element, 507, which assumes phase 1. When 507 assumes phase 1, 505 also assumes phase 1, due to an OR operation, and the loop 504-505-506 also assumes phase 1. At the same time, the loop, 501-502-503, also assumes 10 phase 1. When the :second signal is applied at therinput terminal, 511 assumes a .phase :1 i011 zreceiving thei'nput signal :and makes thephase of 508-509-510 31311383 :1, due to :the :application of :the signal -of phase l from 506 to 511. 0n the other hand, 504 assumes 1phase0 and areturns 504-505-506 to fphaseit), due :10 theapplication of the signal ofa phase opposite .t'o-theqahase'of the input to 504. Similarly, when the .third (signal is applied, 5L2-5l3-5l4 assnmes phase 1 and 508-509-510 returns to phase 0. When the fourth signal is applied, 516-517-518 assumes phase :1 and 51-2-513-5-14= 1'eti1rns tophaseiO. When =the;fiftheignal is apfplzied, 520-52-1=;522 assumes phase 1, and 501-502-503 'is iorcedphase :return to 0, by a feed-back coupling -from:522. the sixth signal is applied,=524-525-526eassumesiphase liand 504-505-506 again assumes phase 1. When the seventh and higher signals and so are applied, similar operations are repeated, except that the last loop, 524-525-526, stays at phase 1. The relation between the dial impulses and the operation of each loop is shown in the Table VI.

TABLE VI loop 504- 505-506 loop 508- 509-510 loop 512- 513-514 loop 516- 517-518 loop 520- coccioose- O OQOHCQOOHO C QQHODQOHOO a OHOOOOWOOO O HQDOOI-IQOQO c b-H-H- HHOQOQO 0 It will be understood that a decimal counting circuit can be made easily. This is done by connecting, to 524, similar circuits to those which follow 504, shown in Fig. 12.

Similarly, any kind of counting circuit can be formed in a simple way, by using parametrons as operating elements.

We claim:

1. A dial impulse register for an automatic telephone exchange, comprising a means for converting the dial current into a high frequency alternating current having one of two phase states, a measuring and detecting means connected to said converting means for measuring the duration of said phase states of the said high frequency alternating current and detecting the dial impulses and lapse of time between the dial current impulses corresponding to one dial digit and those corresponding to the succeeding dial digit, a counting means connected to said measuring and detecting means for counting dial current impulses detected by said measuring and detecting means, and a storing means connected to said counting means for storing the number of dial current impulses thus counted, said measuring and detecting means and said counting means respectively comprising circuits composed of resonators having non-linear reactance elements as a part thereof whereby there can be performed a logical operation utilizing the two possible phase states of the oscillation of said resonators as logical elements.

2. A dial impulse register as claimed in claim 1 in which said means for converting comprises resonator means having a non-linear reactance element for producing a high frequency alternating current, and means for applying said dial current impulses to said resonator means for changing one of the possible phase states of said resonator means to the other possible phase state in response to the make and break of dial current.

3. A dial impulse register as claimed in claim 1 in which said means for converting comprises resonator 1 1 means having a non-linear resistance element for producing a high frequency alternating current, and means for applying said dial current impulses to said resonator means for changing one of the possible phase states of said resonator means to the other possible phase state in response to the make and break of dial current.

4. A dial impulse register as claimed in claim 1 in which said means for counting comprises resonator means for counting the number of breaks in the dial current in the form of a binary decimal code.

5. A dial impulse register as claimed in claim 1 in which said means for counting comprises resonator means for counting the number of breaks in the dial current in the form of bi-quinary code.

6. A dial impulse register as claimed in claim 1 in 12 which said'means for counting comprises resonator means for counting the number of breaks in the dial current in the form of a two out of five binary code.

7. A dial impulse register as claimed'in claiml in which said means for storing'comprise a plurality of ferro-magnetic cores, means for imposing on each of said cores a plurality of high frequency alternating currents each having a phase depending on the numbers dialed, means for superimposing on said cores a current having a frequency one-half that of said high frequency current, whereby each of said cores are magnetized unsymmetrically to one of the two states of residual magnetism inherent in said cores.

i No references cited.

Images