US2513525A - Locking circuit with double signal control - Google Patents

Description

July 4, 1950 E. R. sHl-:NK ET AL LOCKING CIRCUIT WITH DOUBLE sIGN'AL coNTRoL 2 she'ts-sheet 1 Filed Deo. l, 1944 July 4, 1950 E. R. sHENK ETAL LOCKING CIRCUIT WITH DOUBLE SIGNAL CONTROL 2 SheetsA-Sheet 2 Filed Dec.

m m r Y M. n I H @www J 7 ,w m ww ms Patented July 4, 1950 LOCKING CIRCUIT WITH DOUBLE SIGNAL CONTROL Eugene R. Shenk and James R. Weiner, Brooklyn, N. Y., assignors to Radio Corporation of America, a corporation of Delaware Application December 1, 1944, Serial No. 566,129

(Cl. Z50- 27) 6 Claims.

This application. relates to electronic polar relays. Relays of this type have wide application in the radio art. A, particular application is use. of, the same in two-tone telegraph systems. These systems are known also as frequency shift telegraphy systems. n

An object of this invention is improved' relaying of currentsA or potentials.

A more speci'c object of this invention is to provide an electronic equivalent of mechanical' polar relay that is capable of' much greater operating. speeds.

In its, broadest aspect this. invention comprises a locking `circuit somewhat of the. type disclosed.

in. Finch U; S. Patent #1,844,950, dated February 16, 1932. Means: is provided. to subject the control grid of one tube of the. locking. circuit to a. net bias controlled by two independent voltages. Suitable changes are required in these` independent voltages to cause an action in the locking circuit. p

In. accordance with this invention the independent voltages may -be or represent two tones, one of which represents mark` and the other space in a frequency shift telegraphy system.

In describingv this inventionin detail, reference will be` made to the. attachedv drawings. wherein Fig.. 1 shows. by schematic diagram a mechanical polar relay. This ngure is used to make. clear the need of this invention, and also to facilitate description of the invention.

Fig. 2. illustrates by schematic diagram the essential elements of the electronic polar relay oi this invention.

Fig.2a is a chart showing the relation between voltages and the conducting conditions of the tubesv in arrangement of Fig. 2' in the presence of certain control potential conditions which in turn may represent certain signalling conditions, the successive rowsof this chart representing in general conditions at successive instants of time.

Fig.v 3 shows by schematic diagram the' essential. features of an improved two-tone telegraphy system using my improved high speed electronic pol-'ar relay..

The benets that can be. obtained' in manyl applications', for example, in a two-tone telegraph system, through the use of a vpolar relay are well known.

Heretofore, polar relays have been o1." the mechanical type. It is a characteristicv of such mechanical" polar relays' that if the electric current through the coils is reduced tozero, thel contacts. ofthel relay" remain in their l-astoperated rov position. This is usually accomplished through the use of magnetic pole pieces which have sufficient residual magnetism to hold the tongue of' the relay in its last operated position in the absence of any electricl current through the coils.y In order to move the tongue of! the relay, it'is necessary to reverse the current, through thecoil' or -cause current to flow through a diierent coil.

A two-tone telegraph system will be used.. to illustrate how a polar relay can be employed. to advantage to assist in eliminating errors intransmiss-ion. In such a system one tone frequency isl transmitted to indicate marking` intervals, and a dierent tone frequency is used for spacing intervals.

Fig. 1 is a` schematic diagram of a mechanical polar relay. When thisrelay is to be used with a two-tone telegraph system, the circuit is so arranged that Icurrent i1 is controlled by one of the tone frequencies and i2 is controlled by the other tone frequency. Current can be made to flow either when the tone is present or when it transmission circuit such that the spacing tone. no longer was received, thenV i2 would drop toy zero. Under this condition of no current flowing in either portion of.v the. coil', the relay would-remain in its last operated position by virtue of the. mechanical design of the relay and the re' sidual magnetism in the pole pieces.

Suppose now that instead of the spacing toneI dropping out so that neither tone was received.

an interfering, signal', is able.- to cause i1 to flow. Now 4both i1V and i2 Iare flowing simultaneously. If, as is desirable, the magneti'zing. force produced by i1 is the same as that produced. by i2,

the resultant'magneti'zing,'force will become zero.;

Hence, the. netv result of both tones present is the same as neither tone present and` the operated position of the relay does not change.

Consequently, it is apparentv that if. the opere" ated position of the relay is. to be changed, the

spacing tone must go out and' the markingtone must come in. Thisv takes place during normal two-tonetransmission. However, it' is veryim'- probable that spuriousl conditions and signals would cause one tone to' drop out' andv simulcuit of tube V4.

taneously cause another dierent tone of propel frequency to come in. For this reason, a polar relay used in conjunction with a two-tone signalling system provides a very great protection against the recording of false characters.

Among others, it is one purpose of this in- Vention to provide polar operation at much greater speeds than those at which mechanical relays are satisfactory. Fig. 2 is a schematic diagram of a circuit that utilizes the control characteristics of vacuum tubes to simulate the operation of a mechanical polar relay. Since a polar relay must be under control of two separate and independent voltages or currents, two input circuits are provided, one for E1 andA one for E2. As shown in Fig. 2a, E1 is positive during marking signal and zero during spacing signal, while E2 is zero during marking signal and negative during spacing signal.

Tubes or electron discharge systems V3 and V4 are connected in what basically is the familiar locking circuit. The circuit is so named because of its property of locking intoA one or the other of its two stable positions, i. e., whenk the circuit is properly designed, either tube V3 will be fully conducting and the plate current of tube V4 will be Zero or tube V4 will be fully conducting and the current through tube V3 will be zero.

is switched through V3 and cutoi occurs in V4. 1

When a potential such as to cause V4 to conduct is applied to the grid 20, similar action switches the current through tube kV4 and cuts off tube V3.

In addition to the simple locking circuit, tubes V1 and V2 under control of voltages E1 and E2 respectively are included. E1 is applied across a resistance R1 to the grid 49 of V1 which is biased by a source -C. The anode 5I of tube V1 is coupled by resistance R2 into the anode circuit of Vsbetween'Rs and Re. The voltage E1varies from zero (see the chart Fig. 2a), at which tube V1 is biased to cutoff by -Cto a positive value i at which C is overcome and tube V1 conducts.

The voltage E2 is applied across R13 and to the grid 59 of tube V2, the anode of which is cou- This is because when tube V4 is biased at its grid 20 to cutoff, its` anode 24 becomes more positive (potential drop in R4 is less) as does the 4 be conducting with the result that the magnitude of the negative voltage supplied to the grid circuti of tube V4 will be a maximum.

If the tubes, circuit components and supply voltages are properly selected, the net negative voltage on the grid of tube V4 under normal marking condition will be appreciably greater than is required to maintain this tube non-conductive. In fact, it even Will be sufficient that if E1 should fall to Zero (due to a fade out, for example) thereby stopping the ow of current in tube V1 and increasing the positive voltage cou- Apled into the grid 20 of tube V4, the net bias on pled into the grid circuit of tube V4 @by anode load resistance R12. The Voltage E2 varies from 0 at which tube V2, is conductive to a negative value at which tube V2 is cut o.

The plate circuit of tube V1 is interconnected with the plate circuit of tube V3 in such away that it 'can exercise considerable control over the amount .of positive voltage from the source Ebb that is available to be coupled into the grid cir- Tube V2 together with resistors R11 and R12 constitute a voltage divider Ywith two possible settings. One setting is that with tube V2 off. This is the condition of minimum negative bias voltage (magnitude) supplied to V4. When tube V2 conducts, itshunts R12 with the result that the magnitude of' the negative bias coupled into the grid circuitfof tube V4 across R11 is increased. Consequently, thenet gridcathode voltage of tube V4 is ultimately under the control of E1 and E2.

During the reception of a marking signal (E1 positive), tube V1 will conduct and reduce the amount of positive voltage coupled into the grid circuit of tube V4. Simultaneously, tube V2 will the grid of V4 still would be suflicient to prevent space current in this tube. Or, if instead of E1 dropping out, E2 were to come in simultaneously with E1 thereby reducing the magnitude of negative voltage supplied to the grid of tube V4 this tube still would not be able to conduct.

In order to change the sense of the output from the locking circuit, i. e., to cause current to flow in tube V4, it is necessary forboth tubes V1 and V2 to cease conducting. For this to take place E1 must drop to zero and E2 must become nega,- tive. These are two separate and independent changes.

Under these conditions of tube V1 and tube V2 non-conductive, a relativelylarge positive voltage is coupled into the grid of tube V4 from the source Ebb and a small negative voltage is supplied to the same grid from Eea Now the net grid voltage of tube V4 will be Zero or even slightlypositive. When, under these conditions, tube V4 passes current, the positive voltage coupled into tube V3 grid is reduced sufficiently, as a result of the potential 'drop in R4, to render tube V3 non-conductive. When tube V2 ceases to conduct, the positive voltage applied to the grid circuit of tube V4 is increased still farther. Consequently, tube V4 will continue to pass current,' even though either tubes V1 or V2 should conduct individually for any reason.

To return tube V4 to the non-conducting condition, both tubes V1 vand V2 must be made to conduct. This can take place only if voltage E1' becomes positive and E2 goes to zero.

The output may be taken at various points inv the tripping circuit. For example, the voltage or potential across R4 may be taken and used as desired. Y

In the foregoing description the voltages E1 and E2 are referred toas representing mark and space frequencies in a two-tone telegraphy system. Obviously, our improved high speed electronic polar relay may be put to wide use in the radio and allied arts. All that is essential for controlling our polar relay is one voltage Which varies-.from zero to a positive potential, and a second voltage which varies from a zero to a negative value.

vWhen the voltages E1 and E2 represent twotone frequencies such as used in a spaced Wave telegraphy system theymay be derived as i1lu`s`---\ 1 1 are. Separated by the desired band of frequencies.

almanac,

The. separation should at least be sucient to per-mit separation of the same in the circuits following the amplier and limiter stages. A separation of 400 cycles per second has been found satisfactory in one application.

Circuit Vl2 is tuned so as to pass both tone frequencies. The amplifying and limiting circuits of tubes 48, k53, $8 and 18 are substantially similar and are well lknown in the art. Considering tube |8, for example, lthe limiting action of the tube on positive input signals is improved by resistor |9. Resistor 19 causes the grid of tube 18 to see a relatively high resistance generator as compared with the low grid-cathode resistance for `positive grid-cathode Voltage. As a result, voltages that cause the grid to become positive with reference to its ycathode are divided down by ra very high ratio. Further, this grid-cathode resistance is a non-linear function of current which decreases with increasing current so that when the input voltage increases farther positive, the division ratio between the resistor I9 and grid-cathode resistance of the tube increases. The net result is that the positive peaks of input voltage are not present in the plate circuit of tube T8 because the grid cannot become appreciably positive with -respect to its cathode.

The negative peaks of input voltage are not present in the plate circuitof the tube 18 either, due to plate ycurrent cutoii Thus we have essentially symmetrical limiting as well Vas amplication in tubes 4,8, 58,468, andl. Tubes 58 and 68 with resistances 59 and 69 operate like tube 'I8 as described hereinbefore.

The tone voltages representing mark are selected by tuned circuit Mend fed to an amplifier and coupling stage tube .80 while the voltages representing space are selected by a tuned circuit S and fed to an amplifier and coupling stage tube 90. The circuit M is resonated at the mark frequency and the circuit S at the space frequency.

The oscillations of mark frequency are amplied by tube 80 and applied to tube V1 across resistor 95. Tube V1 is a plate rectifier which gives a direct current output proportional to the peak lvalue of the sinusoidal input signal. Resistors 96 and 98 are so proportioned that V1 -is at or beyond plate current cutoi. lResistor 96 and capacitor 99 serve as an R. C. filter to isolate the power supply from signal circuit (with respect to A. C. variations) at rthe point shown. Resistor 94 prevents the grid 49 from going more than slightly positive with respect to the cathode and tends to flatten the plate current peaks of V1.

Assume that V3 is off and no signal voltage appears across resistor` 95. Capacitor |02 will have a `Voltage difference across its terminals equal to Ebb (approximately). `If signal is now applied to resistor `95, tube V1 will conduct on the positive voltage peaks. Capacitor |02 will discharge through tube V1 to some lower value of voltage, the discharge taking place rapidly due to the small internal resistance of tube V1. 'This voltage discharge occurs, of course, when V1 ,conducts due to a positive signal peak. When the input signal ,goes negative tube V1 ceases conducting and capacitor |02 commences to charge back to Ebb. However, it must now charge throughRzi, R5 and Rs whose combined resistance is much larger than the internal resistancel of V1. Thus capacitor |02 will charge slowly (compared with a cycle of signal voltage) and will not reach Ebb before V1 begins to conductagain. y Consequently, the voltage across capacitor |02 remains yatthe lower potential (except for a slight ripple) as long as signals of mark frequency are being received. When the signal ceases, the capacitor voltage again rises to Ebb in a time that may lbe'equal to two or three tone cycles but which is small compared with the length of the shortest signal element received.

Since the tube V1 is biased to cutoi by Een the potential at its anode 5| is positive when the signal drops .out and less positive when signal comes in. Thus lvoltage is coupled into the tripping circuits by resistance R21.

The space frequency voltages selected at S are amplified in tube and excite the cathode |06 of a diode rectifier |08 the anode lof which is coupled to the grid 59 of tube V2. The tube V2 is normally conductive. The `source Ebb produces across resistance ||0 a positive potential at the cathode |06 end of said yresistance so that the diode |08 is biased to cutoff normally. When space tone voltages come in the positive peaks thereof are ineffective. The negative peaks reduce the positive potential on the cathode of the diode so that it passes current which develops across resistance H2 a-negative potential to swing the grid 59 negative -to -cut `off this tube and thereby reduce the negative potential fed :by -Ecc to the grid y20 of tube V4. The condenser ||4 smooths out the rectified pulses in about the same manner in which the pulses are smoothed out by condenser .I 02.

The operation ofthe two-tone systemv will now be described. In this description which repeats the description of Fig. 2 given above, reference will be made to the chartof Fig. 2a.

With nosignal coming-into either channel, tube V1 is cut oiand tube V2 conducts. If a marking signalcomes in, E1 becomes positive, turning tube V1 on, making the grid Vof V4 .less positive. E2 is zero, and Vzcontinues toconduct makingthe grid of V4 very negative. The combination of the two voltages ywill keep V4 shut off and allow V3 to conduct. If the marking signal drops Aout (indicated by dash in the second lineof Ithe first column of the chart) E1 .becomes zero, cutting off V1, making the grid of V4 more positive. E2 Aremains zero and V2 remains on, keeping a high negative bias at V4. In this case the combination of the Voltages keeps the grid of V4 below cutofLand V4 does not conduct. i

If the vspacing signal comes in E1 remains zero, V1 remains cutoff, and the voltage at the grid-ofV4 remains the same due to this act. However, E2 becomes negative, shutting off V2 which raises the potential of the grid of V4 to a point above cutoll". Inthis case vthe combination of the two voltages allows V4 to conduct, thereby shutting Voi V3. If the spacing signal drops out E1 remains zero and the IVoltage at the .grid of V4 due to V1 yremains the same. However, E2 goes to zero andi/2 again conducts, tending to ymake the voltage at V4 more negative. However, it is not sulcient to `drive the grid of V4 below zero, and V4 will remain on.

If a marking signal again comes in, E1 will again go positive, turning V1 on, lowering the potential of the grid of Vi to a point where the combination -of `this voltage and the highly negative voltage due to V2 remaining on turns Vi olf, thereby switching V3 on again.

Noise coming into the spacing channel while a marking signal is being received may turn V2 olf, raising the grid potential of V4. However, this increase in grid Vvoltage is not suiiicient to make V4 .conduct yand V3 Will remain on.

, comes in.

ing channel under these conditions, tube V1 will be turned on, decreasing the positive voltage coupled into the grid of tube V4. In this case, the decrease of positive grid voltage at V4 caused by the conduction of tube V1 is not sui'cent to drive said grid below zero potential, and V4 remains on. The output is taken from the anode of tube V4.

It should be realized that any good type of simple single coil (non-polar) relay may be used in `conjunction with the electronic polar relay of Figs. 2 and 3, for example, connected in series with R4 as shown in Fig. 3, in order to obtain polar operation of a set of contact points.

To those versed in the art, many modifications of the basic invention will suggest themselves.

` As an example, separate circuits could be provided for adding voltages E1 and E2 in a suitable manner and then applying their resultant voltage, or one 'obtained therefrom, to a single point in a locking circuit.

What is claimed is:

1. Apparatus for keying including two electron discharge systems having input and output electrodes cross-coupled and biased to be alternatively conductive and non-conductive, a iirst control tube biased to draw current, a second control. tube Abiased to cutoiT, means for applying alternatively present potentials to said rstV and second tubes to alternately bias said first tube to cutoff and said second tube to be conductive, and a connection between said rst tube and an input electrode of one of said systems, a connection between said second tube and said input electrode of said one of said systems, thel connections and arrangement being such that one of said systems is made conductive only when both of said tubes are biased to be conductive and the other of'said 'systems is made conductive only when bothl of said tubes are biased to cutoi.

r 2.' An electronic polar relay including two tubes each having an anode, a control grid and a cathode, with the anodes and control grids crossconnected, a direct current impedance interconand a fourth tube each havinga control grid and a cathode, means for applying independent signals normally alternately present tothe control grid and cathode respectively of said third and fourth tubes, a connection between the'anode of the third tube and the control grid of said one tube, and a connection between'the anode of the fourth tube and the control grid of said one of said two tubes, the arrangement being such that when one signal only is applied the said one tube has an output of a first value, which is maintained if said one signal drops outor said other signal comes in, and whenthe other signal only is applied the said one tube has an `output of avsecond value, and said output is maintained if said-other signal drops out or said one signal 3. In a telegraphy system, in combination, a

l rst pair of tubes each having a control grid, a

1 anodes cross-coupled so that a change of potential on Lthelanode of one tube changes the ponectin'g the grid and cathode of one tube, athird 8 tential on the grid of the other tube and vice versa, a second pair of tubes one of which is normally conductive and the other of which is normally non-conductive, a coupling between the output of that one of said second pair of tubes normally non-conductive and the anode of one tube of the rst pair of tubes, a coupling between the output of the other tube of said second pair and the control grid of the other tube of said first pair of tubes, and means for applying potentials which appear alternately to the inputs of said tubes of said second pair of tubes.

4. In combination, two tubes having their control grids and anodes cross-coupled by impedances so that when one tube is biased to cut oil' the voltage on the grid of the other tube becomes less negative and vice versa, a rst control tube having its output electrodes coupled with the anode of one of said tubes to vary the potential thereof and as a consequence the potential of the grid of the other of said two tubes, .a second control tube having its output electrodes connected by a voltage divider to the control grid of the other of said two tubes to also control the bias on said grid, means for applying a bias potential to the grid of said rst control tube which varies between a positive'value and zero potential, and means for simultaneously applying to the grid of said second control tube a potential which varies between zero potential and a negative value.

5. In combination, two tubes having their control grids and anodes cross-coupled by impedances so that when one tube is biased to cut o i the voltage on the grid of the other tube becomes less negative and vice versa, a iirst control tube having a control electrode and output electrodes and having its output electrodes coupled with the anode of one of said tubes to vary the potential thereof and .as a consequence the potential of the grid of the other of said two tubes, means for biasing said rst control tube to cutoff, a. second control tube having a control electrode and output electrodes and having its output electrodes connected by a voltage divider to the control grid of the other of said two tubes to also control the bias on` said grid, means for biasing the control electrode of said second control tube by a potential such that it is conductive, means for applying a control potential to the control electrode of said rst control tube which varies between a positive value and Zero potential, and means for simultaneously applying to the grid of said second control tube a control potential which varies between zero potential and a negative value.

6. In a telegraphy system of the type wherein the signals are represented by two potentials of varying magnitude, alternatively present, two tubes each having electrodes including a, control electrode and an anode and having their control electrodes and anodes interconnected by impedances so that when one tube is biased to cut oli,-4

the voltage on the control electrode of the other tube becomes less negative, and vice versa, a rst control tube having a control electrode and having output electrodes coupled directly to the control electrodeof one of said first two tubes to vary the potential thereon in accordance with the signals, a second control tube having a control electrode and .having its output electrodes coupled directly tothe control electrode of said one of said lirst two tubes to vary the potential thereon in accordance with said signal potentials, means for applying one of said signal potentials to the control electrode of said first control tube, and

means for applying the other of said signal potentials to the control electrode of said second control tube.

EUGENE R. SHENK. JAMES R. WEINER.

REFERENCES CITED Name Date Herman 1 Mar. 10, 1931 Number Number 10 Name Date Finch Feb. 16, 1932 Schroter Nov. 28, 1933 Demarest Dec. 12, 1933 Mathes Nov. 6, 1934 Callahan et a1. Aug. 5, 1935 Taylor June 30, 1935 Powell July 26, 1938 Hansell Jan. 2, 1940 Humby et a1 Aug. 20, 1940 Matusita July 22, 1941 Michel May 2, 1944

Images