US2551619A - Electric time constant circuit - Google Patents

Description

May' 8, 1951 E. M. s. MCWHIRTER ETAL ELECTRIC TIME CONSTANT CIRCUIT VFilm Aug. A19, 194'.'

a I lnbenlor Allorne-y Patented May 8, 1951 L ELECTRIC TIME CONSTANT CIRCUIT Eric Malcolm Swift :McWhirter and Roland Harris Dunn, London, England, assignors to International Standard Electric Corporation, New

York, N. Y.

'Application August 19, 1947, Serial No. 769,436 In Great Britain June 21, 1946 Section 1, Public Law 690, August 8, 1946 Patent expires June 21, 1966 4 Claims. 1 Thisinvention relates toelectric signalling or `controlsystems. n l

In many of such-systems .it is necessary to insert -at -anvappropriate ,placer lin .the circuits, a time constant circuit comprising for example a resistance 5in series'and -a -condenser in shunt to smooth out voltagejfluctuations. involves atimerlagin-the systemiand involvesa compromise :between Asatisfactory smoothing of voltage uctuations -and reasonable response -to changes in mean voltage. l

It-is Vthe object of -theinvention to Aovercome this diiculty-and torprovide a circuit arrangement which enables la circuit havingaflong time constant `to be used Atosmooth out slow voltage v'fluctuations whilst at .the same vtime -ensuring a rapid response -to achange -of mean voltage that is-requiredfto be effective.Lv l

According to one of its features the invention consists -of Va variable -electric time constant circuit having a reactive element, a vresistive element consisting of one or Ymore non-linear resistancesanda means of varying the resistance of the said non-linear .resistance or resistances so as Vto vary the time constant of the circuit.

According to anotherof its features the invention consists of a variable electric time constant circuit having -a. vreactive element, a resistive element-consisting of one vor more vnonlinear resistances, a source of vpotential and a means for varying that potential, the said potential being applied to the non-linear resistance or resistances so as to vary the resistance thereof whereby the time constant of the circuit is varied.

-The resistive element of the time constant circuit comprises two resistances and two miniature -rectifiers so arranged that the junction point between the resistances and that between the rectiersare at -the same -potential under balanced conditions. The reverse resistance of the rectiers is connected in circuit so that the time constant is large. When the balance is disturbed the voltage drop -across the resistances alters and the forward resistance of one of the rectiiiers comes .into circuit, thereby reducing the time constant.

According to another of its features the invention consists of a variable electric time constant circuit rhaving a reactiveelement and a resistive element, the latter consisting of ltwo rectiers, opposite poles of Vvvhiclna're connected together andfto the reactive element, whilst the other lends of the said. 'rectiiers arevconn'ected respectively to the positve and negative terminals of -a source of potential, means for varying the potential difference between the junction of the two rectiers and the reactive element on the one hand, and the terminals of the said source of potential on the other hand, the potential between the terminals of the said source remaining substantially constant whereby two alternative conditions only ofthe two rectiers may be brought about namely, on'the one hand both rectiers are biassed in the non-conducting direction a long time constant to the circuit and on the `ther`hand one of the rectiers is biassed .into the conducting condition and the other is biassed into the non-conducting direction giving a short time constant to the circuit.

According to lanother 'of its features the invention consists of an electric signalling or con- -trol system adapted Vto convert -a rate into a quantity having a first condenser which receives periodic increments of charge from 'a vsecond condenser at the rate to be converted, a means of discharging the rst condenser periodically between the times of adding such increments of charge, a time constant circuit the output of which is a quantity varying with the rate to be converted, such time constant circuit consisting of two rectiers and a reactive clement, the rectiers being connected at one end of each to different tappings'on an electric potential divider the ends of which potential divider are connected to a source of potential and joined at the other end of each to a reactive element, the iirst condenser being also connected to a point on the said potential divider, the said rectifiers being connected so as to be capable of being rendered non-conducting by virtue of the potential difference between the said two tappings on the potential divider in relation to the potential at the point of junction of the two rectiers and the reactive element but one rectifier becoming conducting when the rate to be converted varies at a speed greater than a predetermined speed due to a change in the potential o f that point of the potential divider to which the rst condenser is connected, in relation to the potential of the pointof junction of the two rectiers and the vreactive element, thetme constant of the circuit constituted by the two rectiers and the reactive element being reduced when one of the rectiers is in "the ,conducting condition.

The invention :will be 'more readily understood 'from the 'description oi an embodiment 'thereof in relation to the attached drawing 'which is a ii'cui't diagram 'of 'a `device for use Yin a system 4 consisting of R6, R1, R8, R9, and Rl, one point being the junction between R1 and R8, the other being the junction between R9 and RIU.

The junction between rectiers Dl and D2 is connected to C3 which is the condenser element of the time constant circuit, and to the grid of valve V2, connected as a cathode follower.

The manner of operation of the device is as follows:

C2 is charged to a negative potential over impulsing contacts IP, and when these are changed is connected over a potential divider to a point of negative potential to provide a leakage circuit, and a point on the potential divider is connected over a resistance capacity time delay 'circuit to the grid of a cathode follower tube in the output circuit of which is a meter. The leakage circuit stabilises itself for a given rate of adding increments to the condenser and each charge on the condenser just leaks away before the next arrives whenthe increments are being added at the given rate. The time delay circuit has time constant of the order of 20 seconds and smooths out the voltage variations to a straight line value which is transferred to the meter which is calibrated to read the rate of adding the increments of charge to the condenser. By this means the operation of contacts by a kilowatt-hour meter' can be made to cause a meter to register kilo- Watts.

In the drawing a large condenser Cl receives increments of charge via resistance Rl from a' smaller condenser C2 which is itself periodically fully charged negatively over contacts IP in their unoperated position, via a resistance R2.v Y

Condenser Cl is connected across the control grid and anode of an amplifier pentode valve Vl, the screen grid of which is maintained at fthe required potential by a potentiometer consisting of resistances R3, R4, and R5, connected between positive H. T. and earth.

The source of voltage from which C2 is charged, is negative with respect to earth with a voltage which may be approximately equal to the said positive H. T. voltage. I

The anode of'Vl and the opposite terminal of Ci from that connected to Rl, are both connected to a potential divider consisting of resistances RE, R1, R8, R9, and RIU, of which RS is virtually the anode load of VI.

Hitherto, in circuits of this type, CI has been connected via a resistance capacity time constant circuit to the control grid of a valve arranged either as an amplifier or a cathode follower and the output from this valve has been applied to a meter.

It has further been the practice to arrange that the time constant circuit has a time constant sufliciently long to even out fluctuations on the state of charge of Cl so that the meter maintains a steady reading. This has the disadvantage that with slow rates of adding increments of charge to CI, it takes some time for the device to respond to charges of this rate.

When modified according to the invention, the time constant circuit has a condenser element and a resistance element consisting of rectiiiers connected to a source of potential with means for varying that potential so as to vary the rsistance of the rectifiers and the time constant of the circuit.

In the figure two rectiers DI and D2 are connected to two points on the potential divider over, C2 discharges into CI which thus receives an increment of charge for each impulse of contacts IP.

Cl acts as a negative feedback path with respect to the high gain amplier so that the charge on the condenser Cl increases linearly with increasing increments of charge from C2, instead of logarithmically as it would do in the absence ofthe amplifier.

A resistance RM provides further negative feedback to valve Vl and derives its feedback voltage fromthe cathode circuit of V2 to the grid of which the voltage developed across Cl, are delivered.

A leakage path is provided over resistances RT, 8, 9, and I0, for condenser CI, to a point of negative potential and the average state of charge of CI stabilises itself for a given rate of adding'increments from C2, that is to say, for a given rate of impulsing by contacts IP, and each added charge on Cl leaks away before the next arrives.

The resistances R8 and R9 rectifiers DI and D2 and the condenser C3 together form the time constant circuit designed to smooth out the fluctuations in the state of charge of condenser CI as it is periodically connected to C2 by the impulsing of contacts IP, so that a steady voltage is passed to the grid of V2, and the time' constant of the circuit must be long enough to smooth out these fluctuations.

When the rate of impulsing of IP is increased or reduced, the average state of charge of the condenser CI adopts a higher mean value for an increased impulsing rate and a lower mean value for a reduced impulsing rate and a diiferent point of equilibrium is set up. It is required that this changed state of equilibrium shall be conveyed in the form of a changed steady voltage to the grid of valve V2 as soon as possible. For slow impulsing rates, the Xed resistance and capacity time constant circuit hitherto used, tends to delay the assumption of this new steady voltage, which imposes a serious time lag in responding to changed impulsing rates.

The potential divider consisting of R6, 7, 8, 9 and l0 carries a standing current, and the rectiers DI and D2, are connected across R8 and R9.

The standing voltage drop across R8 and R8 is arranged to be slightly greater than the value of each increment of voltage added to condenser CI and it appears as a reverse voltage across rectifiers DI and D2 which are so poled as to be ina high resistance state, a state which for brevity is referred to as the non-conducting condition though it is realised that the rectiers are never (and indeed must not be) completely non-conducting.

To understand the operation of the device R8 and R9 may be considered as a single resistance.

The voltage across the rectiers DI and D2 is mainly governed by the vvoltage drop across'R, R9, but the voltage across either of the rectiers singly, also affected by the potential difference between the point of its attachment to the potential divider R6, 1, '8, 9 and 10 land the point Y'of its attachment 'to C3 Aand the 'grid of uV2.

This potential difference is affected by a change in the 'potential at any point of vthe potential divider R6,`1,8, 9, III.

The changing state of charge of CI as it is periodically charged and discharged effects such a change in the `potential of the said potential divider and the potential applied from time to time across each of the said rectfiers is a fluctuating potential.

stable, the potential -of the grid :of V2 becomes stable at an average value between the extremes of iiuctuation of the voltage 'applied by C'I to the rectiiiers via the said potential divider and in fact tends to vfollow the average vpotential of the contre point of RB, R9. The vstanding potential across R8., R9 is chosen so that the fluctuations, as CI is charged 'and discharged 'at a steady rate, do not reach the point where either of the rectiers DI and D2 becomes conducting.

The potential divider R6, 1, 8, 9, IU, combined with condenser CI, constitutes a source of potential for biassing the rectiers in that the former provides a steady and the latter a iluctuating potential superimposed thereon, the uctuations of the combined potential being within such limits that both rectiiiers remain non-conducting when the rate of adding increments oi charge to CI is steady.

The system is then balanced to the particular rate of impulsing of contacts IP and under these conditions, the smoothing time constant of the circuit is governed by the high resistance of the two rectiers together with the capacity of C3.

When the impulsing rate of contacts 1P changes, however, and the rate of adding increments of charge to CI alters, the fluctuation in the voltage delivered to the said potential divider by CI will occur about a raised or lowered mean value according to Whether the impulsing rate is raised or lowered, and the potential of the various points on the said potential divider will b e changed relative tothe potential of the junction of DI, D2, C3 and the grid of V2, unless the rate is changing so slowly that the potential at this junction has time to adjust itself to the new level despite the long time constant of the circuit.

On the peaks of the fluctuations, therefore, the potential across one of the rectifiers between its point of attachment to the said potential divider and to the grid of V2, will overcome the standing potential across R8, R9 and cause this rectifier to become conducting.

This rectifier will therefore be in a low resistance state (which, for brevity is herein referred to as the conducting condition) and the time constant of the circuit including this rectifier and C3 will be shortened considerably.

The circuit arrangements which bring about a change in the general level of charge of CI in response to a change of the impulsing rate of contacts IP, when the change is suijiciently rapid, as explained above, constitute a. means of varying the potential applied to the rectiers DI and D2 by the potential divider R6, 1, 8, 9, III, in combination with CI.

At the time therefore, when the uctuation of the potential from CI makes one of the rectiers conducting, there will be a rapid change of the voltage on the grid of V2 in the direction corresponding to the change in the impulsing rate of IP.

When the fluctuation of 'the potential of C'I passes its peak, the rectifier 'which was conducting will pass to the non-conducting condition again till the next similar peak, and if the potential of the grid of V2 has not completely adjusted itself to the changed impulsi-ng rate, the rectifier will again become conducting and the potential on the grid of V2 will again change.

:This "will take place repeatedly until the potential of the grid of V2 has adjusted itself to the new impulsing rate, whereafter the iiuctuation in the potential of CI will no longer produce a potential difference between the connection of the rectifier to the said potential divider and to -the-grid of V2, sufficient to 'overcome the standing bias potential across R8, R9, and both rectifiers will remain non-conducting, giving a slow time constant to hold the potential of vthe grid of V2 steady.

The arrangement therefore gives a -high time constant at a steady -rate of impulsing of IP, and thus eiectively smooths out fluctuations in the voltage delivered to the grid of V2 but allows of a rapid response to a change in the rate of impulsing.

In the drawing, CI is shown as connected to the junction of R8 and R9 but it will be clear that CI can be connected to various points of the said potential divider according to the convenience of circuit design, so long as iiuctuations in its state of charge are arranged to cause iiuctuations of the potential of various points on the said potential divider and particularly the points of connection thereto of the rectiiiers DI and D2. The connection of CI to the junction of R6 and R1 has been found to be successful and such a connection is shown in dotted lines in the drawing, the condenser CI being re-designated CI1.

In the drawing, a meter M is shown connected via a variable resistance RIZ to a potentiometer consisting of RII and RIS in series, constituting the cathode load of V2. RI I is variable to adjust the zero setting of the meter.

This meter records the rate of impulsing of contacts IP by measuring the output of the cathode :ircuit of V2 which follows the voltage on the grid V2.

RIZ enables the response of the meter to be calibrated.

The charging time constant of CI may be adjusted to suit dierent impulsing rates and different values for CI and if this time constant is shortened beyond certain limits it may be advisable to add a further condenser element to the smoothing time constant circuit constituted by rectiers DI and D2 and condenser C3.

This may be effected by connecting a condenser C4, where shown in dotted lines in the drawing, from the junction of R1, R8, and rectifier DI to earth.

It is sometimes of advantage to reduce the heater current of VI to reduce grid current at times when the grid of the valve is driven positive.

In the drawing, a resistance RI5 is shown in series with the cathode heater supply leads, for this purpose.

We claim:

l. A variable electric time constant circuit having a long time constant in its normal condition of operation for smoothing out slow iiuctuations in an input signal and a short time constant for ensuring rapid response to change in said input signal comprising a reactive element and a resistive element connected together, said resistive element 'including two rectiiers, opposite poles 4of said rectiers being connected together and to said reactive element, a source of direct potential, a potentiometer connected across said source,

-plying said input signal to a point on said potentiometer whereby rapid changes in the input signal cause unbalance of the voltages aboutthe bridge formed by said potentiometer and said rectiiiers and a biasing of one of the rectiers into 'a conducting condition, thereby reducing the time constant of the reactive and resistance elements. 2. A circuit according to claim 1 in which said reactive element comprises a condenser.

3. A circuit according to claim 1 in which said input-signal-applying means is connected to a point on said potentiometer intermediate said positive and negative points.

8 4. A circuit according to claim 1 in which said rectiers are connected to positive and negative points designed to bias said rectiers so that they remain in the non-conducting condition except when the frequency of the uctuations of the input signal exceeds a predetermined value.

ERIC MALCOIM SWIFT MCWHIRTER. ROLAND HARRIS DUNN.

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

UNITED STATES PATENTS Mortlock -1 July 2'?,

Images