US2434894A - Apparatus for converting pairs of time modulated pulses into pulses of variable duration - Google Patents

Description

Jan. 27, 1948 D. M. AMBROSE 2,434,894 APPARATUS FOR CONVERTING PAIRS OF TIME MODULATED PULSES INTO PULSES OF VARIABLE DURATION Filed June 24, 1943 2 Sheets-Sheet l JOLWZF 1/412 FWGEE aurflur PULSE F5670 M'F T F M F F M M F M Inventor Demof Min fimrox.

By I" v Jan 27, 1948.

D. M. AMBROSE 2,434,394 APPARATUS FOR CONVERTING PAIRS OF TIME MODULATED Y PULSES INTO PULSES OF VARIABLE DURATION Filed June 24, 1943 2 Sheets-Sheet 2 an FINGER l5 ms: J. FEED INPUT T F/G.4. F M F j F M F M r f 5oosEco/vas.|

Inventor Dermof fil/n flmMse,

Patented Jan. 27, 1948 OFFICE A. non.

APPARATUS FOR CONVERTING PAIRS OF TIME MODULATED PULSES INTO PULSES F VARIABLE DURATION Dermot Min Ambrose, London, England, assignor to Standard Telephones and Cables Limited, London, England, a British company Application June 24, 1943, Serial No. 492,078 In Great Britain September 26, 1941 Section 1, Public Law 690, August 8, 1946 Patent expires September 26, 1961 2 Claims. (Cl. 25027) The present invention relates to circuit arrangements for producing variable duration pulses from time modulated pairs of constant duration pulses.

In a known type of electrical communication system the intelligence to be communicated is transmitted as a modulation of the time duration of the pulses occurring at equal intervals of time, the variation in the duration of the pulses, i. e., the time modulation, being proportional to the amplitude of the wave form representing the intelligence at instances equally spaced in time. Such a system is known as a pulse modulation system. In a development of the pulse, modulation system, marking pulses, which are in themselves of constant duration are transmitted to mark the beginning and end of a time modulated pulse. These marking pulses are herein referred to as finger pulses.

In this specification the following definitions will be used:

A finger pulse comprises a short surge of voltage or current, the duration of which should preferably be less than ten per cent of the repetition time or of that order.

A train of finger pulses is a series of finger pulses occurring at successive intervals of time.

A fixed pulse is a finger pulse which occurs at successive equal time intervals.

A moving pulse is a finger pulse-whose time of occurrence with respect to a fixed finger pulse is variable.

The leading edge of a pulse is the edge formed by the initiating change of voltage or current which forms the pulse from its initial state to the peak value of the pulse.

The trailing edge of a pulse is the edge formed by the change of voltage or current which terminates the pulse.

One application of the invention is for the translation of intelligence in a double pulse system for which purpose it is necessary to produce waves having a square wave form from a train of finger pulses in such a manner that the leading and trailing edges of a wave correspond to the two finger pulses marking the leading and trailing edges of the time modulated pulses,

Another application of the invention is in obstacle detection or distance measuring systems utilising the transmission and reflection from an obstacle of electrical pulses. The time elapsing between the transmission of a pulse and reception of its echo is a measure of the distance. In one system pulses of duration equal to the time elapsing are produced from the transmitted and received pulses and used to give an indication of the distance required.

It is the object of the present invention to provide circuit arrangements for producing waves having a square topped wave form corresponding to the pairs of fixed and moving finger pulses in a train of finger pulses.

Accordingto the present invention circuit arrangements for producing variable duration electrical pulses from time modulated pairs of constant duration electrical finger pulses comprise a variable voltage device to which the pulses are applied to control the voltage, one of the pulses of a pair producing a sudden voltage change in one direction to a new value and the other pulse returning the voltage suddenly to its original value.-

In one arrangement according to the invention, a train of positive finger pulses is fed to a generator of oscillations of sawtooth wave form, herein referred to for brevity as a sawtooth oscillator, in such a manner that the oscillator is triggered by each fixed finger pulse of the train. Finger pulses are formed from the resulting sawtooth wave by limiting the output to the positive tips of the wave and these finger pulses are fed to one of the control grids of a double stability" circuit. The other control grid of the double stability circuit is fed by a corresponding train of finger pulses which are of the same polarity as the pulses produced from the sawtooth wave form and may be of the same polarity as, or of reversed polarity, and derived from, the original finger pulse train. The output from one anode of the double stability circuit is then in the form of the square waves required.

In another arrangement for producing square wave forms from a train of finger pulses, the pulses are made to charge a condenser across which a gas discharge device, for example one such as those known under the registered trademark Thyratron is connected, and by adjustment of the grid bias of said device, it is possible to set the voltage at which it will strike and initiate gas discharge, to be in excess of the voltage caused by charging of the condenser by the first pulse. As the second pulse continues to charge the condenser, the gas discharge device strikes and provides a discharge path for the condenser which remains so discharged until the arrival of the first finger pulse of the next pair to begin another cycle.

Practical embodiments of the invention will now be described with reference to the accompanying drawings in which Fig. 1 shows circuit arrangements of an embodiment utilising the generatonof oscillation pf sawtooth wave form;

Fig. 2 shows curves explanatory of the operation of the circuit shown in Fig, 1;

Fig. 3 shows circuit arrangementsof an embodiment utilising a condenser and gas discharge device; and

Fig. 4 shows curves explanatoryof t e opera.- tion of the circuit shown in Fig. 3.

Referring now to Fig. 1 of the drawings, atrain of fixed and moving positive finger pulse of a constant amplitude are fed across terminal I and 2 to the control grid of valve VI and through a small condenser H to the control grid of valve V2. The valve V2 is a .'Ihyr.atron, orother li-ke gasdischarge device, and thecircuit associated with it is.of the conventional sawtooth oscillator type. Theoperation of V2 is synchronized by the pulses impining Ql'laitS controligrid. The freerunning speed of the oscillator .is adjusted .by the resistance l3 to be very nearly .the same as that of the fixed pulses, .sothat the resultingout- ,putis of a freguency identicalto that of the fixed pulses. The condenser H and grid leak resistance 12 of V2 difierentiate attheir junction the input finger pulses so that shorter alternate .positive and negative pulses corresponding .respectively to the leading and trailingedges of themput finger pulses are appliedtolVZ. The vertical ,edgeof the sawtooth waveform, may then be synchronised with either of these positive or negative pulses and thus may be made to coin- .cide with either the leading ortrailing edg of .theinput finger pulses.

The operation is 'found to be stable within limits depending upon the duration of thefinger pulses. When this duration becomes very short and the pulse repetition frequency ;is not increased correspondingly, the circuit .hasa-tendency only to synchronise with the leading edge of the-finger pulse since there will, underthese conditions, belno sensible change in .the charging voltage time .of the condenser [4 across the Thyratron V2 in relation to that between the leading and trailing edges of-the fingerpulses.

The sawtooth waveform from the oscillator V2 is fed to the control grid d5 of a highdm- .pedance pentode V3. This gridis biased negativelyby the resistance I l in such a :manner ,as .to cause the valve to accept only -a; small;posi;ti ve tipof the sawtooth wave. The output -f r o n the anode .19 of V3 is thus anegative pulse thewidth of which may be varied by the bias control ,l 'l.

;tl '1e, grid This pulse is fed to oneend (in thiscase of valve VG) of a double-,stabilityflc c prising valves Vt and -V 5-;which;are nedium impedance triodes and are interconnected by resistancesfii and23 a shown. It will beseenirom thecircuit that if the control grid of V i goes neg t e 1 4 ceas to P s an d ret a hence its anodegoes positivecausing the control grid oi .Viito go i ore pcsitite and llcnce JV; to draw increased anode current. Thiscauses the anode of Vito become ,rnore rregative.and ,hence other words, whennegative pulses are impinged condenser 4 9.

onthegriu-of Vdthe anode of V5 will be more negative, and vice versa.

The-priginalifinger pulse train after being reversed in polarity by valve V1 is fed to the control grid of-V5 and the generated pulse from the saw-tooth wavejsfed to the control grid of V4. {ljhisjatterjsof slightly longer duration than the cqrresponding pulse (i. e., fixed pulse) in the original traimthe duration being adjusted by bias control 1"! a already stated. Thus the doublestability circuit is actuated by the generated pulse on one grid whilst the moving pulse only is.eife ctive on the. other grid.

The output of ,the desired waveform is taken between terminals andJl.

Referring to Fig. 2 of the drawings, curve a shows thefixed fin ers pulses F and ,the moving finger pulses ,M. The leading .edge of a ,pulse is indicated .L.

Curveh. indicate the [waveform obtained .at, the outputterminals and .4 after reversal insign. It will beobservcdthat the .leading edge L of .the waveform coincides with th ,leading edge Lof the corresponding fixed fingerpulseahd the-trailingedge T of ,the waveform coincides with the leading edgeL of the corresponding moving finger pulse.

Referring now to Fig. 3 of :the accompanying {drawings which shows .the circuit arrangements :utilising a .condenser and gasdischarge device, the train of positive fixed .and moving finger .pulses are fed across terminals 1 and}. These pulses are thus impinged on to the cathode ll of diode Di .andanode 1. 3 of diode D2. Diode D! is -for,D. ctrestoration purposes. only, i..e., any negative .voltag-e appearingon its cathode ,-l:l causes current to flow from its anode 15 thus .by passing all negative .pulses. This .valve may beused in thecircuit'butis not essentialto the workingof thecircuit. The positivepulses pass through D2 from its anode i3 to cathode t1 and charge-the Thecharge .is .held on the con- ,denser by .the high impedance which D2 offers to the discharge of the condenser.

Connected across the condenser .lil {is a fljhyratron or other ,like .gas discharge .device Vi whosebias isadjusted bythe resistanc 2| to such a value that the striking-voltage necessary is greater than the voltage produced by the charge onthecondenser-JQ due tothe first pulse and less than the voltage producedby thecharge .due to two successivepulses.

The value .of the condenser i5 depends upon .theinput pulsewidth i, e., the time. during which thesurge. occurs. and alsolon the amount of power which is available .inthe pulses ,For exarnple, if the pulses used be of .a '25 rnicrosecond duration a this combined with a medium impedance triode ieedingthe eirsnit with-pulse the optimum va .Q (the san a i is and :t be .ePPrQ .maicly mic ra Th w ve of the fing r Pu e andih pulses across the output of valve VI are shown in curves a and b of Fig. 4, respectively. As in Fig. 2 the fixed pulses are marked F and the moving pulses are marked M.

Thus before a fixed finger pulse F arrives the condenser I9 is discharged and at the reception of a fixed pulse F the charge and therefore the voltage across the condenser suddenly rises as shown on the left hand side of a pulse, in curve 19. Then during the interval between finger pulses F and. M the charge remains substantially constant. On the reception of a finger pulse M the charge of the condenser suddenly increases as shown by the peak on the right hand side of a pulse on curve 1). During the charging up of the condenser on the reception of a pulse M, the voltage exceeds the striking voltage of the gas discharge device VI which ignites and provides a path for the discharge of condenser I9, so that the voltage across the condenser drops suddenly as shown by the right hand side of a pulse, curve I), Figure 4. Thus the pulses in curve b are of the same durations as the durations between the corresponding fixed and moving finger pulses.

The final output from this circuit arrangement has to be kept at a high enough impedance to prevent the condenser discharging between the pulses by more than 10% of its initial charge. There are three possible paths of discharge, viz.:

(1) Through D2 and DI,

(2) Through the Thyratron VI.

(3) Through resistance 3| and 33.

In case (1) the impedance of D2 is to megohms and in case (2) the impedance of VI when it is not in its ionised condition is 30 megohms since it passes 10 micro-amps at 300 volts.

Thus case (3) is the limiting factor and in a practical case 3 megohms was found to be sufiicient for the pulses used which were of a repetition time of 2000 per second and a duration of '75 microseconds.

The pulses in the output of VI may be tapped gill 1 a suitable point of the series resistances Owing to the fact that the output waveform is not perfectly rectangular the top and bottom of the waveform has to be limited.

The limiting action maybe obtained in one valve of medium impedance but a higher impedance valve V2 is preferred in order to utilise the shorter grid swing and the sharper cut-off. The pulses obtained in a tapping from 3 I-33 are applied to the control grid 23 of the valve V2. The resistance chain formed by resistance 35, and parallel connected resistances 2| and 29 may be replaced by only two resistances in series across the H. T. supply. The reason for using such a chain as that shown for supplying bias should be obvious to those conversant with the art both for the proper operation of the Thyratron VI and for V2, the latter being the limiter valve. The amplitude of the signal output from VI is arranged to be considerably in excess of the grid swing voltage of valve V2. Thus the valve V2 is run from the overload point to the cut-01f point of the anode current and so the top and base of the square wave-form are limited. The accepted amplitude limits are shown between the broken lines 25 and 21 curve b Fig. 4 and these limits may be moved between the limits of the amplitude of the pulses fed to V2 by means of the bias control resistance 29.

What is claimed is:

1. Arrangement for producing variable duration electrical pulses from a source of time modulated pairs of electrical finger pulses each said pair comprising a fixed pulse and a moving pulse, comprising a double stability electronic device including a pair of anode-cathod space discharge paths and a pair of control electrodes one in each of said paths, connections from the anode of each said path to the control electrode of the other path, an output circuit connected across one of said space discharge paths, means for producing a train of electrical pulses of substantially the same periodicity as said input fixed finger pulses but of longer duration, mean for Synchronizing said train of electrical pulses with said fixed finger pulses, means for applying said train of electrical pulses to one of said control electrodes and means for applying the input finger pulses from said source to said other control electrode.

2. Arrangement for producing variable duration electrical pulses, from a source of time modulated pairs of electrical finger pulses each said pair comprising a fixed pulse and a moving pulse, comprising a double stability electronic device including a pair of anode-cathode space discharge paths and a pair of control electrodes one in each of said paths, connections from the anode of each said path to the control electrode of the other path, an output circuit connected across one of said space discharge paths, means for generating sawtooth waves of substantially the same periodicity as said input fixed finger pulses, means for applying said input finger pulses to said last-mentioned means to synchronize said saw-tooth waves with said input fixed finger pulses, a limiter device, means for applying said saw-tooth waves to said limiter device to cause it to pass only small tips of the saw-tooth waves, means for converting said small tips and said input finger waves to like polarity, means for applying said small tips to one of said control electrodes, and means for applying said input finger pulses to the other of said control electrodes.

DERMOT MIN AMBROSE.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 2,266,401 Reeves Dec. 16, 1941 2,092,887 Luck Sept. 14, 1937 2,086,918 Luck July 13, 1937

Images