US2434965A - Time modulation pulse system - Google Patents

Description

Jan. 27, 1948. R. B. SHEPHERD 2,434,965

` TIME MODULATION PULSE SYSTEM Filed Aug. 1l, 1943 3 Sheets-Sheet 1 F/Gx' ian. 27, l948. R. B. SHEPHERD 2,434,965

' y TIME MODULATION PULSE SYSTEM Filed Aug. l1, 1943 5 Sheets-Sheet 2 WAL/f fa me Jan. Z7, 1948. R. B. SHEPHERD Y 2,434,965

TIME MODULATION PULSE SYSTEM Filed Aug. ll, 1943 Z5 Sheets-Sheet 5 ATTwrA/'mf Patented Jan. 27, 1948 UNITED STATES Par Applioation August 1.1, 1943, serial' No. 4stats In Great Britain September 11, 1942 16 Claims'. (Cl. 179-'1-1-7115) The' present invention relates to electrical pulse transniissiony systems and is` principally concerned with arrangements' for obtaining' time'- modulated pulses.

According to the invention, there is" provided an electrical pulse transmission system compris'- ing means for' supplying respectively each of two saw-tooth wave generators with synchronising alternating voltages of the same frequency but differing in phase, means for' varying the phase difference in accordance' with a modulating signal', and means for combining the saw-tooth waves froml the two generators' to produce a series of' regularly repeated rectangular pulses timemodulated in accordance with the` signal.

The invention may' alternatively comprise a system for deriving time-modulated. electrical pulses from two saw-tooth wave generators', comprising' a variable phase' splitter' controlled. by a modulating signal ariel` adapted' to transmit two separate synchronising voltages having' the same frequency and a vari-able phase diiierence from a sinusoidali oscillator tothe two generators respectively, and a` mixing device for' subtractingv one saw-tooth wave from the other to produce rectangular pulses.

According to a diierent aspect, the invention consists in' the method of time-modulating ai series of regularly repeated electrical pulses which comprises; synchroni'sing two saw-tooth generators. with alternating! voltages having' the same frequency and'. diierentlph'ase; varying the phase difference' accord-ingI to a modulating signal and combining,v the outputs of the'. generators to' ob#4 tain rectangular pulses.

The invention.: will'. be explained' with reference to the accompanying. drawings in which:

Fig.. 1 shows wave-formY diagrams' usedfor explaining. the. invention;

Fig.l 2 shows-',ablock schematic diagram of the principle of the invention;

Figs. 3` and 4 show schematic circuit' diagrams'A of' avariable phase splitter according; to the'invention; and- Fig. 5 shows. a schematic circuit diagrami ot a'- mixer accordingl to the invention.A

Fig. 'sliows a schematic-circuit-diagram of a system according. to Fig.v 2- inwhich the phase splitter shifts the phasevofboth component waves derived-V from the. oscillator.

Fig.' Tshows a systemwith twophase splitters utilizing directly heated thermistors.-

Fig. 8 showsa system operating. on. the prin cipl' off' Fig. 4 but with' two phase.v splitters inf' stead of one.

It is well known that rectangular pulses 'may' be' obtained b'y combining a series ofsawaooth waves with a, similar series of waves l'iavingv the same amplitude and' frequency, which havebeen' inverted and displaced in phase. This i's1 shown in Fig. l where the' curve A represents a series of saw-tooth voltage waves having a period T and amplitude V. Curve' B; shows the invertedwave's' which have been displaced by a' time' t with' re` spect to thev waves A. f these two series oi waves are added together in a` suitable' mixer the rrec'- tangular pulses C are produced, having a' totalv height V, and period T. The' positive portions" have' a voltage proportional' tov t V/ T and a width T-i, and the negative portions' have a voltage proportional to (T-t) V/T and a width t. Buy wou known arrangements' unidirectional' pulses' of either' sign, having width T'-i or i', as den sire'd, maybeA derived from the pulses C'. It' may' be noted' that the' phase di-ierence F between the waves A and B is 2xt/T. l i v x Referring. to Fig. 6, the phase splitter PS has two parts PS1 and PS2, each of which comprises' a bridge network having two condensers and* two indirectly heatedthermistors. The two thermistors in PS2 have temperature coeicients opposite to those of the two in PS1, those in PS1 having a positive temperature coefficient and those in PSZ- anegative temperature coeiii'cient, or vice versa. A sinusoidal oscillator 0^ iscon-'-A nectedY to the input diagonal terminals ofv PS1`- The output diagonai terminan of the networkI PS1 areconnected1,through wave lter WF-Ll if desired'to terminals a;-a, leading. t'o` tle' terminals of sawtooth generator SG1. The out-f putdiagonal terminalsl`v ofnetwork PS2 are" con; nected through.l wave lter WFZ if desired; to

terminals. b b which in turnlead to the iripiit` of sawtoothrgenerator- SGz. The output terr-nil' nalsV ofy sa-wtooth generators SG1v and' SGE; a e connected ina mixing circuit M liketha'tsliowi? in Fig. 5, orin an' equivalent. arrangement; infsuoli a, way that a difference voltage appe'arsa-li terminals 5i,- 6- as` explained' in connectienlv with' Fig. 5.-

In Fig. 6 each component of the wave fromoscillator Q is:r shifted inA phase' by substenttally:y they saine amount but' opposite' direction'sl so thatftheresultant pulses have leading and trail ingedges shifted 1in' opposite directioiisI with spect to the center line of the pulse. The center line of the pulse has a phase position or time position corresponding to the phase of the unshifted oscillations from O.

Thus, if the signal voltage increases, raising the temperature of the thermistors in PS1 and PS2, for example, the resistances of the thermistors in PS1 may be assumedto increase while those in PS2 decrease. If this change causes a phase lag in the output of PS1, then the same change causes a phase lead in PS2. Similarly, when the signal voltage from M decreases, the resistances in PS1 will decrease while those in PS: increase, and a phase lead is produced in the output of PS1 with a phase lag in PS2. Y

Assuming that A in Fig. 1 represents the output wave from SG1 and B the output wave from SGz when there are no signals from M, then A will be shifted slightly to the right for a lag in PS1 due to signal from MS and B will be shifted slightly to the left for a lead in PS2 due to the same signal from MS. The distance t will become smaller and the distance (T-t) larger so that the (positive) pulse is widened. If the signal changes so that A is shifted to the left and B to the right, t becomes larger, (T t) becomes smaller, and the pulse is narrowed.

The width of the unmodulated pulse is dependent on the initial phase difference between the outputs of PS1 and PS2, that is, the phase difference when there is no input signal from M. In order that the leading and trailing edges of a pulse shall shift equal amounts, to give a symmetrical double-modulation effect, a given change in signal at M should be made to produce equal and opposite phase shifts in PS1 and PS2.

There are various obvious modifications of the circuit of Fig. 6, such as connecting the heating coils of the thermistors in parallel instead of in series, or adding any desired auxiliary resistances in the heating coil circuits. If the heating coils are omitted entirely, the modulating current is supplied at terminals [-2 along with current from oscillator O; in this case the wave lters WF-I and WF-2 are necessary to eliminate the modulating current from the output. These modifications have been described in connection with Figs. 3 and 4, and apply as well to Fig. 6.

Fig. 7 illustrates a system utilizing directly heated thermistors instead of indirectly heated thermistors. As indicated, the modulating current is supplied at terminals l, 2 along with synchronizing current from the oscillator, and the heating coils r1 and r2 are lacking. In other respects the system is closely similar to that utilizing directly heated thermistors. The operation of the system is believed to be obvious.

Fig. 8 shows a two phase-splitter sytem each bridge circuit of which operates on the principle of the circuit shown in Fig. 4. That is, the capacity of one arm of each bridge is varied by varying that component of the capacity of the arm which is under control of the tube, VT-l or VT-2. The modulating current, suitably rectified to provide variable bias for the grid of the tube, varies the amplification factor of the tube and so varies the effective tube capacity to unbalance the bridge. This produces phase shift in the bridge output current. By applying biases of reversed signs to VT-l and VT-2 the phase shifts are produced in opposite directions, giving rise to the production of two sawtooth waves shifted in opposite directions. When combined in the mixer M the two sawtooth waves produce doubly modulated rectangular pulses in a manner analogous to that described in connection with Fig. 6.

The pulses so obtained may be time-modulated by varying the phase difference F in accordance with a modulating signal which varies t (or T-t) without changing T. The manner in which this may be done is shown in Fig. 2, in which an oscillator O supplies alternating current at frequency I /T, and having preferably a sinusoidal form, to a variable phase splitter PS. This is arranged to supply two similar saw-tooth generators SG-I and SG-Z with synchronising voltages of frequency I/T and differing in phase by F, which is controlled by a modulating signal applied at MS. The saw-tooth generators SG--l and SG-Z may be of any suitable type, and the outputs areconnected'to a mixer M where the two saw-tooth Waves are combined to form the (n pulses. vThemixer may contain the means for inverting one of the waves' and for making the resulting pulses unidirectional. The output of the mixer is connected to the apparatus (not shown) in which the pulses are to be used.

The means for varying the phase difference F in the phase splitter PS is not shown in Fig. 2 and may take a number of forms two of which are shown respectively in Figs. 3 and 4. The arrangement shown in Fig. 3 takes advantage of the special properties of thermistors, which are thermally sensitive resistance elements.

Thermistors have been in use for some years and are characterised by a temperature coeicient of resistance which may be either positive or negative and which is moreover many times the corresponding coeiiicient for a pure metal such as copper. This property renders thermistors particularly suitable for a variety of special Iapplications in electric circuits.

Various different materials are available for the resistance element of a thermistor, these various materials having diierent properties in other respects; as one example, a resistance material having a high negative temperature coeiicient of resistance comprises a mixture of manganese oxide and nickel oxide, with or without the addition of certain other metallic oxides, the mixtures being suitably heat treated.

Thermistors have been employed in two different forms: (a) known as a directly heated thermistor and comprising a resistance element of the thermally sensitive resistance material provided with suitable lead-out conductors or terminals, and (b) known as an indirectly heated thermistor comprising the element (a) provided in addition with a heating coil electrically insulated from the element. A directly heated thermistor is primarily intended to be controlled by the current which flows through it and which varies the temperature and also the resistance accordingly. Such a thermistor will also be eiected by the temperature of its surroundings and may therefore be used for thermostatic control and like purposes with or without direct heating by the current flowing through it. An indirectly heated thermistor is chiefly designed to be heated by a controlling current which flows through the heating coil and which will usually, but not necessarily, be different from the current which flows through the resistance element, but this type of therrnistor may also be subjected to either or both of the types of control applicable to a directly heated thermistor.

More detailed information on the properties of thermistors will be found in an article by Y y the saw-tooth waves.

G. L. Pearson in the Bell Laboratories Record, Dec. 1940, page 106.

In this speciiication any resistanoes not actually referred to as thermistors are ordinary constant resistances whose values do not depend appreciably on the current iiowing through them.

I Eteferring to Fig. 3, the phase splitter PS comprises a bridge network comprising two condensers C1 and Cz, and the resistance elements R1 and R2 of two indirectly heated thermistors T1 and T2, arranged in opposite arms as shown. The synchronising alternating voltage is supplied at terminals I and 2, and is applied through a wave lter WF--I to the saw-tooth generator SG-I (not shown in Fig. 3). It is also applied t-o one pair of diagonal terminals of the bridge network, the other pair of terminals being connected to the saw-tooth generator SG-Z (not shown), through a second wave lter WF-2.

It is well known that the phase of the synchronising voltage supplied to WF-2 will diier from that supplied to WF-l by an amount depending on the values of the impedances in the bridge arms, and the value of the phase difference can be changed by suitably changing any or all of these impedances. In Fig. 3 the heating coils r1 and r2 of the thermistors are shown connected in series to a pair of input terminals 3 and 4 to which is supplied a modulating current which Varies in accordance with some signal voltage. The thermistors should both have temperature coeiiloients of the same sign which, however, can be positive or negative as desired. As the signal voltage increases, for example, the temperature oi both the thermistors will be raised and their resistance will be changed in the same direction. This will change the phase of the current supplied to the wave filter WF-Z; and will in turn change the phase diierence F between the two saw-tooth waves and also the width of the pulses in the manner already explained. Similarly, when the signal voltage decreases, F will be changed in the opposite direction. It will be evident that by changing the sign of the temperature coecients of the thermistors, the changes in F will simply be reversed. V

The two filters WF-I and WF-2 should be designed to pass substantially only the synchronising frequency, in order to eliminate any extraneous frequencies which may be present and which might affect the proper synchronising of The filters are, however, not absolutely essential to the working of this arrangement.

It will be understood, of course, that the heating coils r1 and rz could be connected in parallel instead of in series, and any other auxiliary resistances could be added in the heating coil circuit in any desired way.

By a modification of Fig. 3, the two thermistors could be of the directly heated type, in which case the heating coils r1 and r2 shown in Fig. 3 would not exist, and their connections would accordingly be left out. The modulating currents are in this case supplied with' the synchronising currents at terminals l and 2 andflow with them through the bridge, variably heating the thermistors and changing their resistances. The wave lters WF-I and WF-2 are in this case essential in order to exclude the modulating currents from the saw-tooth generators.

The arrangement shown in Fig. 3 is only suitable for relatively low frequencies on account of the time lag in the response of the thermistors. For high frequencies, the arrangement of Fig. 4

may be used, which depends for its operation on the property of thermionic valves known as the Miller eilect, which arises from the capacity between the anode and the control grid. Itis well known that this capacity provides a feedback path in the valve and operates also to cause the effective capacity measured between the control grid and the cathode to be increased by an amount depending on the amplicationfactorof the valve and on the` anode load. lInformation on this eiect may be found on page 97 of Modern Radio Communication," vol. II, by J. H. Reyner, published by Pitman (1940).

In Fig. 4 the bridge network consists of two resistances Ra and R4 and two condensers Ca and C4. The condenser Cs is shunted by the control grid-cathode circuit of a thermionic valve V of the variable amplification type, which is caused to vary the total effective capacity in the Ca arm of the bridge in a manner to be explained. The bridge corners are connected to the wave lters WF-I and WF-2 and to the terminals l. 2 as in Fig. 3.

The valve V is provided with a grid leak comprising two resistances Rv and Ra in series, and

R'z-i-Rs should preferably be large compared with the impedance of the condenser C: at the synchronising frequency. Thev anode is supplied with potential from a battery or other source S through a load resistance R9, and a conventional cathode resistance R5 shunted by a condenser C5 (or any other suitable arrangement) may be used for biasing the control grid.

The modulating signal currents are rectified by any suitable means (not shown) and the rectiiied output is applied through the terminals 3 and 4 to the resistance Rv.

The control-grid voltage so obtained will vary in' accordance with the signal and will vary the amplication factor of the valve, and thus also the eiective capacity connected across C3 in the manner just explained. This will varylthe phase diierence F in accordance with the signal, as V l before.

The sense of the variation may clearly be reversed by reversing the sign of the rectified voltage supplied to terminals 3 and 4.

In this arrangement the wave lters WF-l and WF-2 are necessary in order to ensure that the modulating currents are excluded from the generators.

It will be understood that although the valve'y sired. Various other arrangements are clearly possible by which any or all of the impedances may be varied in' accordance with the signal.

Moreover, the type of bridge network shown is only one possible type of phase changing network that might used. Such a network could include inductances instead of or in addition to any of the impedances shown, and the impedances might be arranged in other ways; and further, it need not be a bridge network.

It will be noted that in Figs. 3 `and 4 the phase splitter comprises only one phase changing network, which is connected between the terminals said two generators respectively each of said bridge networks comprising two condensers connected respectively in opposite arms of the bridge and two ordinary resistances connected in the remaining arms, one pair of diagonal points of the bridge being connected to the oscillator and the other pair to one of the generators, a thermionic valve having a variable amplication factor shunted across one of the condensers and means to control the control grid-cathode circuit of said thermionic valve in accordance with a signal.

10. A system according to claim 9 having means for controlling the ampliiication factor of the valve including a circuit for applying to the control grid a rectified voltage derived from the signal voltage.

11. A system according to claim 9 further including a circuit applying the modulating signal to vary the amplication factors of the two valves simultaneously in opposite directions.

12. A system according to claim 1 in which thel means for supplying the two alternating voltages comprising each generator is preceded by a wave iilter adapted to pass substantially only the synchronising voltage.

13. The method defined in claim 16 in which the saw-tooth waves are combined by subtracting one from the other.

14. A system for deriving time modulated electrical pulses from two sawtooth wave generators comprising two sawtooth wave generators, a sinusoidal oscillator, means including a phase splitter for deriving from the output of said oscillator two separate sychronizing voltages having the same frequency but different phase, means for applying said separate voltages to the two sawtooth wave generators respectively, a modulating signal source, means to control the said phase splitter by said source so as to vary the phase difference of said synchronizing voltages by said modulating signal, and a mixing device for subtracting one of said sawtooth waves from the other to produce rectangular pulses.

15. An electrical pulse transmission system comprising two sawtooth generators, means for supplying to said two sawtooth generators respectively synchronizing alternating voltages of the same frequency but of different phase, said means including a sinusoidal oscillator and a variable phase splitter, a source of modulating signals and means for controlling said phase splitter by modulating signals from said source so as to vary the phase of at least one of said voltages, whereby the phase diierence of said two voltages is varied by said signal source.

16. The method of transmitting a signal through the medium of regularly repeated electrical pulses which comprises producing a variable signal current, producing two alternating voltages having the same frequency but differing in phase. causing said two voltages to produce two separate sawtooth waves having the same period but separated in phase by the phase difference of said alternating voltages, controlling the magnitude of the phase difference between said alternating voltages in accordance with the variations of said signal current, and combining said sawtooth waves to produce rectangular pulses of variable duration varying in accordance with said phase diierence.

RONALD BRADLEY SHEPHERD.

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

UNITED STATES PATENTS Number Name Date 2,250,819 Wolf July 29, 1941 2,226,459 Bingley Dec. 24, 1940 2,086,918 Luck July 13, 1937 2,019,481 Applegate Nov. 5, 1935 2,287,174 Heising June 23, 1942 2,227,815 Toulon Jaan. 7, 1941

Images