US2375950A - Frequency divider - Google Patents

Description

Patented May 15, 1945 FREQUENCY DIVIDER Kurt Schlesinger, West Lafayette, Ind., asslgnor to Radio Corporation of America, a corporation of Delaware Application. March 4, 1942, Serial No. 433,289

1 Claim.

This invention relates primarily to circuits adapted for use in frequency reduction systems as may be carried out by oscillating systems having a natural oscillatory frequency which is either equal to or slightly lower than the desired reduced frequency. Such circuits find use, for instance, in television apparatus and particularly so for developing the synchronizing signal pulses which are transmitted with the signal or video pulses, or such apparatus is particularly useful for developing the well known blanking signals.

In the prior art, circuits of the blocking oscillator type or the multi-vibrator type in its various forms belong generally in this general group of circuits. However, it has been found in practice that circuits of the general type known in the prior art are not always completely and sufficiently reliable for the above specified uses, because of the self-oscillatory character of the circuits heretofore known. The primary reason for the generally unsatisfactory experience with the prior art circuit is because of the fact that no matter whether or not a synchronizing signal is present to control these circuits, it has been found that the circuits go through oscillations at or near the natural frequency and immediately prior to reaching the so-called firing point they are especially and highly sensitive to the smallest voltages which could serve to trigger them. Such triggering voltages are frequently due to no more than the stray fields of nearby oscillators, or may be due to other undesirable synchronizing signal voltages or noise, from which it is desired that the circuit under investigation be completely free. Also, it is not feasible to set up start-stop systems with the oscillatory type of frequency dividers.

Therefore, in the present invention it is a fundamental object to design and develop a circuit particularly adaptable for use in frequency reducing systems which does not oscillate freely but stands by indefinitely in an attending position until the controlling or synchronizing signal impulse arrives and triggers the circuit at precisely the desired time for its operation.

It is a further object of the present invention to provide a circuit of the character hereinabove referred to in which the circuit, after responding to a controlling signal or impulse, is induced to perform only one period or cycle of oscillation which is preset to the desired length of time. After performing the one cycle of oscillation, the circuit then returns to an attending position, as before, where it is ready to pass through another cycle of oscillation immediately upon receipt of another controlling impulse.

In one form, this circuit may be designated a triggered oscillator system in that the oscillations produced are holly and completely dependent upon the rece t of a controlling impulse, and under the receipt of the controlling impulse the circuit temporarily acts as an oscillator.

Another object and advantage of .the invention is that of producing a circuit for accomplishing the results hereinabove named, wherein the circult is made as free as possible of any circuit elements which would tend in any sense to place it in a self-oscillatory state.

A further object of the invention is to provide a circuit which is adapted to function as a frequency divider and yet be relatively simple in its arrangement and construction and still be positive in its operation and easy to control.

Other objects and advantages will become apparent from a reading of the following specification and claim in connection with the drawing.

In the drawing,

Figure 1 represents diagrammatically a series of received synchronizing impulses, and diagrammatically represents the time at which the triggered oscillator will pass through one cycle of oscillation;

Figure 2 represents schematically one form of circuit for accomplishing the aforesaid results; and,

Figure 3 represents a multi-stage frequency divider circuit of the general character hereinabove shown by Figure 2.

Considering now the drawing, it will beseen from Figure 1 that a series of synchronizing impulses s, each of a time duration t and of an amplitude e0, are adapted to be received at the input terminals ll of the frequency divider (as in Figure 2). If the system is so constituted that a plu rality of impulses s are caused to pass by before the oscillator system is again triggered, then it will be evident that in order to perform a fre- This voltage will then serve to cut off the action of an amplifier at some predetermined time until it has leaked away through a resistor, and such an action may occur at any time between the n and n+lst impulse, so that the system shall then be induced to the new state of operation again by the n+lst impulse. From what has been shown. it can be seen by referring to Figure 1 that in order to avoid firing before the n+1st impulse, the residual voltage to be built up across 4 such a condenser must be at least equal to er=n-eo, where n is the ratio of desired frequency reduction.

With a simple rectifier circuit (for instance, one in which a resistor is connected across the anode and cathode of the rectifier and a condenser is connected serially with, say, the anode 4 of the rectifier) it is impossible to obtain such an amount of residual voltage. To the contrary, a

television. Accordingly, such a system as that herein described immediately above is unable to perform frequency reduction, and instead every' impulse which arrives is transmitted by it without omission.

In order to obtain the desired frequency reduction effect, it is apparent, in principle, that what might be done is to connect an amplifier having a predetermined gain so that it is energized from the diode and caused to feed its energyback to the condenser. Basically under such conditions of a positive impulse e is applied to the plate of the diode through a condenser. Then, an amplifier tube should be able to turn out a negative impulse to its cathode simultaneously, and such an impulse would have a greater amplitude represented by g-ee, where g represents the gain in the tube, and, as a result of this phase reversal and amplification between amplifiers, an adequately increased residual voltage across the coupling condensers might be obtained.

Following the above reasoning and referring to Figure 1, it will be seen that the basic principle required to perform frequency division is thus set forth, and it will be noticed that the system does not perform oscillations because it is nothing but an amplifier system operating in combination with a time circuit (RC) across its input circuit.

Referring now to Figure 2 in the light of what has been hereinabove explained, impulses of the general character and wave shape shown by the impulses s of Figure 1 are applied to the terminal point I I and arranged to be supplied to energize a vacuum tube l5. The general wave form of these impulses is indicated in Figure 2 by the wave form shown adjacent to the input terminal point. The tube I5 is connected as a resistance coupled amplifier and is preferably a tube of the 6N7 type, so that it consists of two independent triode sections within a single envelope. Whatever impulses are applied to the terminal H are then fed by way of the condenser H to the control electrode I9 of the tube IS. The first stage of the tube comprising the cathode 2|, the control electrode l9 and the anode 23 acts as a phase inverter and an amplifier, and the second stage of the tube I5 comprising the cathode 25, the control electrode 21 and the anode 29 acts simply as a cathode follower stage. In normal operation the first stage of the tube is connected so that its grid or control electrode member I9 is connected to the resistor 3|, which, in turn, connects between the cathode bias resistors 33 and 35, of which the latter connects to ground at 31. The anode element 23 is connected to the highv voltage terminal line 39 by way of a resistance element 4|, and, similarly, the anode element 29 connects to the high voltage line 39 by way of the resistor element 43, which element may under some conditions, as will hereinafter be explained, be eliminated where desired, as, for instance, when no output energy is derived at the terminal point 45. To feed the energies from the first stage of the tube l5 to the second stage there is provided a connection by way of conductor 41. and potentiometer 49 in such a way that the potentiometer is tapped at the tapping point 5| to provide suitable voltage for controlling the control electrode 21 of the second half of the tube. The end of the potentiometer 49 removed from the anode element 23 is also connected to ground at a point 31.

Where it is desired to derive negative output impulses, rather than the positive impulses obtained at terminal 45, a connection may be made at the terminal point 53 so that the voltage there derived represents the potential at the cathode tapping point 55.

While the tube |5 has been shown and illustrated as being, for instance, of the so-called 6N7 type, it may readily be formed and operated as two completely separate tubes and, where desired, the tubes may be of the tetrode or pentode type, as well as the illustrated triode.

Now, to explain what is believed to be the operation of the circuit of Figure 2, it should be borne in mind that the purpose of the invention is fundamentally that of providing a very rigid control of the triggering action as compared with the phase of the exciting impulses. Broadly speaking, this action is such that the second half of the tube l5, which includes the portion consisting of the cathode 25, the control electrode 21 and the anode 29 acts as a cathode follower stage and is for the purpose of providing power amplification. Thus, the second half of the tube I5 has the effect when connected, as herein disclosed. with the first half of the tube comprising the cathode 2|, the control electrode l9 and the anode 23, of carrying the condenser charge (that is, the residual charge in the condenser "1 beyond cut-off by a long way, so that the rate of return of the condenser H to a sensitive condition is controlled by the magnitude of the resistor 3|.

Considering the circuit in the foregoing light and to enter into a more detailed explanation of the operation, it will be seen that normally the first stage of the amplifier is blocked beyond cutoil. In this case, and to refer to the specific illustration herein used, the second stage has a grid bias of the order of about half th positive voltage in the supply line 39, or in other words, about 100 volts with the positive voltage assumed at 200 volts, as shown. The common cathode voltage above ground comes reasonably close to the grid or control electrode voltage, and may be chosen to be of the order of about volts. Therefore, no plate and no grid current will flow in the first stage of the tube until the grid voltage at least reaches, or even exceeds, the pre-set cathode-to-ground voltage. However, if this happens, an impulse of at least 10 volts (that is, the voltage across the resistor 33, for instance) is developed, and the first stage of the tube l5 will act as an amplifier and phase inverter, and the second stage of the tube will act as a cathode follower with its grid and cathode carried down toward ground potential in a negative peak of the following amplitude: e g-e0, where 9 equals the gain of the first stage. Consequently, the condenser I1 is strongly charged up .by grid current and the first stage is again blocked oil. Therefore, an adjustable multitude of triggering impulses will pass without effect until the n+1st impulse again drives the grid or control electrode 59 above the cathode-to-ground bias potential.

The degree of frequency reduction is set by the tuning of the condenser I'l and/ or the resistor 3| and depends also upon the input voltage. As is apparent from the showing of Figure 1, the frequency reduction may decrease from n to nl if the input voltage is increased, but the order of frequency reduction must always be an integral number, as, for instance, the integral number 5 as shown by Figure 1. According to the showing of Figure 2, negative impulses at the reduced'frequency appear at the cathode tapping point 55 and may be derived in the polarities shown at the terminal 53. Positive impulses appear at the terminal point 45, which represents the potential at the plate or anode element 29. If it is desired to couple the circuit into the next succeeding stage, then it is frequently desirable to derive the control voltage at the terminal point 45 (as shown by Figure 3 particularly) in which case the resistor 43 is the desirable part of the circuit Otherwise the resistor 43 may be omitted.

Further, in operating the circuit hereby disclosed, it is desirable that the resistance 33 be so adjusted as to prevent the first grid or control electrode l9 ever reaching the cut-off level by itself, and, accordingly, the resistance 33 is approximately 10% of the value of the resistor 35. If desired, the resistance elements 33 and 35 may be in the form of a potentiometer with the fixed terminal here shown constituted by the tapped point on the potentiometer, but it can be seen that if the biasing herein disclosed were not provided, the system might start into oscillation at approximately the period of the circuit comprising the condenser element I1 and the resistor 3|, even if no triggering or control impulses were present at the input terminal I I.

Referring now to Figure 3 of the drawing, the arrangement disclosed is substantially that of Figure 2, with provision made for connecting such frequency divider as it is shown by Figure 2 in series in order to obtain frequency division of high order. Such a use, for instance, might be found in television synchronizing signal generators. In such cases, the coupling circuit comprising the condensers l1, l1, l1 and the resistors 3|, 3|, and 3| act as the time constant circuits and have to be tuned to giv the desired frequency thereto, whereas the connection of the amplifiers is essentially the same at all stages. The positive input pulses are shown as being produced across the cathode circuit of a blocking oscillator tube 51 (of well known form). The anode element 59 of the oscillator 51 is energized from the power supply line 39 and arranged to feed back energy to the control electrode 6| by way of transformer primary 63 feeding energy into the transformer secondary 65 which con-' nects through the condenser 61 to the control electrode 6| on the one hand, and to ground at 69 on the other hand. The cathode output resistor ll of the tube 5'! is of the usual type to derive the desired potentials. It should be noted that under the arrangement shown by Figure 3, the anti-oscillation resistance element 33 of Figure 2 is omitted, as the operation of the system is continuous and each stage develops positive pulses across its second plate to trigger the following stage.- By varying the value of the plate or anode resistors 43, 43', as shown, the trigger voltage may be adjusted to some extent, and thereby the order of frequency reduction likewise varied. In the reference to Figure 3, the like parts to those shown by Figure 2 have like designations, and like parts in succeeding stages also have like designations, except that a prime or a double prime number is used, as the case may be.

In cases where the arrangement of Figure 3 is used in the synchronizing signal generator of a television transmitter, the frequency at which the pulses from the blocking oscillator tub 51 are developed may be 31,500 cycles per second (assuming a 525 line picture repeated at 30 frames) and 60 fields per second. Then, by following the usual frequency reduction provided in the synchronizing signal generator for deriving the field frequency of 60, it can readily be seen that the output as derived at the terminal point 45 may be one-seventh the frequency of the oscillator 51, or 4,500 cycles, which will serve as the input to the second stage comprisin the tube l5, for instance, and. the tube l5 then may dividethe frequency by five so that the impulses appear at the terminal point 45' at 900 pulses per second. Similarly, the stage 15" may divide that frequency by five so that the pulses at the terminal point 45" appear at pulses per second, and then, by the addition of one stage which reduces the frequency by a third, the output will be developed at the 60 cycle rate corresponding to the number of picture fields per second, and also corresponding to the frequency of the usual power supply mains. Similarly, to derive the line frequency impulses occurring at 15,750 pulses per second (for the same type raster) an arrangement of the type shown by Figure 2 will be energized from the blocking oscillator 51, but the frequency reduction will be of the order of so that the blocking oscillator 51 assumed to develop output energy at 31,500 pulses per second the line frequency pulses occurring at 15,750 pulses per second can readily be determined.

From the foregoing it is evident that the duration of the output pulses of reduced frequency would be substantially the same as that of each input pulse, but where an integrating circuit comprising the resistor 43 and the shunting condenser 40 is used, the output desired may be effectively lengthened, if desired. It is very desirable, however, to maintain the percentage of the pulse duration constant, as compared to the repetition period, rather than their absolute duration, as is the case in a television synchronizing signal generator, for instance. To vary the length of the pulse, a condenser is arranged to shunt the tube output resistor 43, as shown, and by a suitable choice of the resultant time constant, the pulse output may be lengthened and adjusted, as desired, and is lengthened in proportion to the decrease in frequency. Further illustration than that herein incorporated is believed to be unnecessary.

It is, of course, apparent that many and various modifications in the circuit herein disclosed may be made without departin from the spirit and scope of the disclosed invention, and therefore, I believe myself to be entitled to make and use any and all of the modifications hereof which fall fairly within the spirit and scope of what is here disclosed.

Having now described my invention, what I claim as new and desire to secure by Letters trolled electrical discharge paths provided between separate cathode and anode elements ot the said thermionic means, a connection between the cathode elements of the pair or separate discharge paths to place the cathode elements nor-- mally at substantially like potential, a resistance path 'connected between the cathodes oi the said discharge paths and a point of substantially .iixed potential, means for producing normally a predetermined current fiow through one of the discharge paths to provide a voltage drop inthe said resistance path connected between the cathode and the said point of fixed potential, a connection between the anode of the other of said discharge paths and the grid of said first discharge path for interrupting current flow in the said first discharge path during periods of current flow in the second of the discharge paths, a capacity element connected to provide a controlled bias voltage upon the second of ,the discharge paths tocontrol the current flow therein. a resistance means connected to an incharge the said capacity means in accordance termediate point on the cathode'resistance to with the voltage drop across theflrst named resistor between the connection point and the point of fixed potential so as to prevent the bias of they second discharge path from ever. exceeding cutoi! value7-in the absence of input-pulses, input terminal connection'means for supplying received signal pulse energy of substantially constant periodicity to the said capacity, whereby following the receipt oi.a predetermined number of signal energy pulses the second of said discharge paths is maintained substantially instantaneously conductive and the capacity means is discharged-

Images