US2358297A

US2358297A - Blocking oscillator

Info

Publication number: US2358297A
Application number: US348877A
Authority: US
Inventors: Alda V Bedford
Original assignee: RCA Corp
Current assignee: RCA Corp
Priority date: 1940-07-31
Filing date: 1940-07-31
Publication date: 1944-09-19
Anticipated expiration: 1961-09-19

Description

p 1944- A. v. BEDFORD 2,358,297 BLOCKING OSCILLATOR Filed July 51, 1940. 2 Sheets-Sheet l p Q v *B'VOL 77465- ZSnventor VOL 7 14 715 056/4 40702 FREOl/E/VCY Sept; 19, 1944.

A. v. BEDFORD BLOCKING OSCILLATOR Filed July 31, 1940 2 Sheets-Sheet 2 Patented Sept. 19, 1944 BLOCKING OSCILLATOR Alda V. Bedford, Collingswood, N. J., assignor to Radio Corporation of America, a corporation of Delaware Application July 31, 1940, Serial No. 348,877

4 Claims.

This invention relates to improvements in television systems and, more particularly, to an oscillator for synchronizing the operating action at a receiving station with that at a transmittingstation.

In television receiving systems using a cathode ray tube to produce optical images from the received signal energy, a deflection system for causing a saw-tooth current wave at a. predetermined line scanning frequency is used to de, flect the ray in one direction, for example, horizontally. There is also a second deflection sys= tem for causing a saw-tooth current wave at the desired framing frequency to deflect the ray in another direction, for example, vertically.

For the purpose of synchronization, an oscillator, such as the blocking oscillator shown in the Tolson, et al., Patent No. 2,101,520, granted December 7, 1937, generates sharply peaked output voltages which are utilized to drive the deflection systems in synchronism with the deflection systems at thetransmitter.

. While'these systems have been found to lend themselves to satisfactory operation, a distinct operating disadvantage thereof resides in the "fact that the frequency of oscillation of the oscillators previously used is controlled by the applied plate voltage. Relatively small variations in supply voltages, therefore, are effective to change the frequency sufiiciently to throw the system out of synchronism.

There are many reasons why a more'stable oscillator is desirable. For example, a more stable oscillator will remain in step with synchronizing pulses of lower amplitude, interference for a given location will have a constant amplitude relative to the amplitude of the synchronizing v signal and therefore by theapplication of less signal to the deflecting circuits there is less interference signal applied. This results in a smaller difference in scanning line length which difference is due to the irregular amplitude of the synchronizing pulses when an interference sigof, he O ca ima e- A oth very m ant d anta e gained by the use, of a more stable synchronizing oscil- Since azsine wave is f not desired, it is not nput voltage wave is generated by employing an -,lato-r that it ispossible to dispense with the manual controls of the existing system for maintai in the os lla r in s h o sm t the t a smitted. s n h on zi pu s possible to use a tuned circuit'oscillator without great complication, but it is necessary to provide a generator producing sharply peaked voltage pulses with good frequency stability.

There are substantially three things which govern the frequency of an oscillator; namely,

the tube characteristics, the circuit constants, and the voltage supply. If the frequency of the discharge device and its associated circuit are not alfected' by temperature changes or other like disturbances, it remains that the only item affecting the frequency is the oscillator circuit voltage supply.

According to this'invention'for the deflection of the cathode ray, a sharply peaked outirnproved oscillator circuit whose frequency response remains substantially constant during plate supply voltage vai'iations. This is accom-v plished by biasing the control electrode with a positive potential derived from the anodefsupply voltage. A change in positive control electrode bias voltage has the opposite effect-on the oscillator frequency, as does the change-in the anode potential in the same direction. This results in a cancelling of change, so that the oscillation generator frequency remains substantially constant.

With the foregoing in mind, the primary object of this invention is to provide an improved oscillator circuit for generating sharplypeaked output voltages which maintains its required frequency of oscillations independent of occurring variations in the supply voltages over a range usuallymet with in practice.

Other and incidental objects of the invention will be apparent to those skilled in the art from a readingof the following specification and an inspection of the accompanying drawings in which Figures 1 to 4 are circuit'diagrams showing different forms of this invention, and

Figures 5, 6, 7 and 8 are explanatory diagrams relating to the operation of this invention.

Referring to Fig. 1, the oscillation generator includes a discharge device such as a thermionic tube I having a cathode 3, a control electrode 5 and an anode I. A transformer 9 provides inductive coupling between the anode 1 and control electrode 5. Theprimary I I of the transformer 9 is connected in the'anode potential supply circuit and secondary I3 is connected to the control electrode through a coupling condenser "the condenser a e path for the electrical charge stored in the coupling condenser I5. A potentiometer I9 is inserted across the anode voltage supply source. and its variable tap 2| is connected to and impresses a positive potential on the control electrode of tube I, which positive potential is directly proportional to the anode supply voltage. The synchronizing voltage from a television receiver, and for example one such as shown and described in U. S. Patent No. 2,137,039, is fed to the circult through an auxiliary transformer winding,

23. The output of the oscillator is taken from the anode circuit through the coupling condenser 25. "I

The operation of the oscillator circuit will now be explained with reference to Fig. '7. Time is represented along the abscissa and plate voltage represented by E; and Epl, plate current repre sented by Ip and Ipl, control electrode bias represented by E19, E91 and the control electrode potendraws current.

compared to the resistance between the control electrode 5 and the cathode 3 while control electrode is at a positive potential with respect to the cathode 3 and the time elapsing between t1 and t: is relatively small with respect to the time taken for each complete cycle.

The range of voltage represented from c to Egl is the voltage stored across the condenser l5 because of control electrode rectific'ation'during the time when the control electrode is positive and The negative charge stored in condenser l5 at time its and represented by voltage 0 is beyond the cut-off point of tube I in a negative direction, so that no anode current flows from time tsto t4 and this charge is allowed to leak off through resistance I! as represented by tiallrequired to out off the anodecurrent is represented by Eco. At the beginning of 'a cycle, for example, at time tithe control electrode 5 is at out off voltage and the anode lstarts to draw current through winding ll of transformer 9 causing an induced voltage inftransformer winding 13. The polarity of the transformer winding l3 with'respect to winding 1 l is such that the control electrode 5 is then driven more positive. An increase in positivebias on the control electrode 5 causes a still further increase in anode current. Due to the fact that the transformer 9 presents an inductive load to the anode current in its winding H the voltage impressed on the controlelectrode 5 tends to lead the anode current in winding II so that when the anode cur-' rent approaches saturation the polarity of the potential induced in winding 13 reverses and causes the control electrode potential'to return rapidly to its axis which has shiftedin the nega-' tive direction due tothe accumulated charge'on the' condenser I5 caused by control electrode rectification.

Anode current is limited at time tz-b'ecaus'e' of the saturated condition of tube l. The anode current limiting action of thisitube condition is due to the fact that the negative charge of the electrons flowing from the cathode 3 to the anode 1 tends to repel any further electron emission from the cathode 3 sothat when a large number 'ofelectrons are in the space between the cathode "3 and the anode Ldue to a heavy current flow,

this negative charge is suflicient to prevent any further increase in anode current. It is also true that the electron emission of the cathode 3 is limited by its physical characteristics.

-' When the positive control electrode potential ,to the very rapid decrease in plate current at the start of the cycle and the control electrode potential therefore returns to a value c corresponding to the negative charge stored in the condenser l5. During this cycle of the control electrode voltage the negative charge stored by I5 is not appreciably affected by resistance I! because the latter is relatively high the control electrode potential curve between is and 14 which takes the form of a logarithmic curve'approaching Egl asymptotically. The rea-' son for the logarithmic decay of voltage, is that at point c'the stored potential is large and thereforeaheaviercurrent flows, as the potential due to the stored chargein condenser l5 decreases the current decreases, hence the charge decreases. less rapidly near the end of the curve.

It can be seen that the positive bias Egl from the potentiometer tap 2| increases the rate oi. decay of the storedpotential because the curve approaches this'potential Egl instead of the oathode or zero potential. However, the control electrode potential curve intercepts the bias poten-- tial value Eco, representing the cut-off voltage of the tube 1|, before allthe storedpotential has been dissipated through resistance I1 and. at; this point'anode currentbeginsto flow again and thecycle is repeated. V v

The frequency of the recurrence of the pulses of..theoscillator can thereforebe adjusted by;

. changing the value. ofresistance l1 in order to change the rate-atowhich ,the'negative charge leaks 'oflf condenser l5 toallow the tube l toagain draw current; V.

.,The time ii to is is determined primarily by the natural period of the windings H and I3 of the transformer 2| and their associated circuits and the anode draws current only during this time as shown by, the curve marked Ip.

Since the load circuit of the anode is inductive the setting up of a counter-E. M. E, due to the current in the inductance, results in a derivative. type of voltage wave in the output circuit. Such a Wave is represented by curve Ep.

. In actual operation, the time in to is covers; only a very small portion of the total time it; to t4. representing a complete cycle of the oscillation generator. Time ii to if; hasbeen extended in the graph to more clearly show the shape of the: curve. As previously explained, this portion of the curve tito' t depends forits shape and time:

.on the characteristics of the coupling transformfer, control 'electrode,'jjand coupling condenser.

trol electrode approaches the cut-off value of the tube IL Although the controlelectrode potential curve approaches a value of-bias voltage represented by Egl, it will be" readily understood that, long before the energy stored in' condenser I5 is allowed to leak off completely, the control electrode potential will reach its cut-off value and cause the flow of anode current and consequently the start of another cycle as represented at $4.

The resistance I1 is adjusted so that'the charge stored in the condenser at is leaks off at a rate such that the control electrode potential approaches closely to the cut-off point at time t4. At this time iii, a sharp synchronizing pulse E9 is impressed upon th control electrode circuit to raise the potential on the control electrode past the cut-off potential so that at time t4 conditions will be such that the anode can again draw current to start the next succeeding operating cycle.

The time required for the discharge of condenser I5 is determined by the ratios-existing between the value of resistance II, the capacity of the condenser I5 and the positive control electrode potential applied by the potentiometer I9 through its variable tap 2|. An increase in positive control electrode potential increases the rate of discharge of condenser I5.

In actual practice, the anode supply voltage B+ is likely to change in value, and if oscillators of the type shown in the prior art are used, any decrease in anode potential causes a corresponding increase in oscillator frequency because of a smaller amount of stored potential in condenser I5 and thus a shorter time interval between is and t; as shown by the dashed lin in Fig. 8. This can be further explained by the fact that under conditions of lower anode voltage such as Epl the control electrode potential reaches a voltage a1 rather than voltage a. The reason for this is evident in view of the fact that a decreased plate voltage Epl is accompanied by a corresponding lower anode current Ipl, this in turn reduces the control electrode peak voltage to an causing less control electrode rectification voltage to be stored across the condenser I5.

Referring to Fig. '7, it follows that the control electrode bias is reduced proportionally because this bias potential is derived from the anode voltage through the potentiometer 9 and its tap 2I. This lower control electrode positive bias is represented in Fig. '7 as Egl. Because of the lower control electrode rectification voltage stored in condenser I5 under this condition of lower anode potential, the control electrode potential returns to point 01 at time is. At this point the storedvolta-ge begins to leak off through resistance I1. However, this ime the logarithmic voltage curve approaches asymptotically th line representing the control electrode bias Egl. So that by properly adjusting the circuit constants the two curves representing different B+ supply'potentials may be made to intersect the cut-off value of the tube I, represented by Eco, at substantially the same point.

In Fig. 5 is shown the effect of a change in anode potential Ep on the oscillator frequency. It will be noticed that between certain practical limits the oscillator frequency decreases at a substantially uniform rate during an increase in anode voltage Ep. It is also apparent from the curve Eg that the oscillator frequency will increase with an increase in positive control electrode bias voltage. The increase in oscillator frequency due to a change in control electrode bias is much more rapid for a given change of potential than is the case of the change due to a change of anode potential. Therefore, by causing the control electrode positive bias to increase in proportion to any increase in anode potential, a curve results which is represented in Fig. 6

in which the oscillator frequency remains substantially constant over a practical range of 31+ supply voltage. V r

Fig. 8 shows the operation of such an oscillator as disclosed in Fig. 1 without the application of a positive bias to the control electrode. 7 It will be noted that the'control electrode potential returns to potential 0 and-approaches asymptotically to a line represented by zero control electrode voltage. However, in the case of a lower anode potential Epl the control electrode voltage returns to or and returns asymptotically to the same reference of zero control electrode bias as does the logarithmic curve'beginning at 03, a greater negative control electrode potential. It necessarily follows that each of these two curves will intersect the control electrode cut-oil potentialEco at different points, resulting in a different oscillatory frequency under conditions of different supply potential. The B+ supply potential may be low enough to cause the control electrode potential to reach the cut-off voltage prior to the time of thesynchronizing pulse Es and cause a start of another cycle before t4. On the other hand, the anode supply voltage may be great enough to cause the control electrode bias to be such at time 124 that the synchronizing pulse Es wont raise it to the cut-off value and, consequently, the oscillator won't be triggered at time it and the frequency will drop down.

Another form of this invention is shown in Fig. 2 in which the storage condenser I5 is placed in parallel with the leakage resistor I! and this combination is placed in series with the transformer winding I3 which is, in turn, connected to the control electrode 5. The operation of this circuit is substantially the same as for the one shown in Fig. 1, and previously described. The

fact that the bias potential from potentiometer tap 2| is not inserted across the storage condenser does not in any way alter the rate of discharge of control electrode rectified voltage through resistance II. In the case of Fig. 2, the synchronizing input" voltage is impressed across a resistance 25 connected in series with the control electrode circuit so that there appears a synchronizing potential across the reand its associated circuit causes a slight change.

in the shape of the synchronizing pulses.

If we examine the control electrode circuit in Fig. 1, we find that condenser I5 feeds the synchronizing pulses to resistance II, which has a relatively low impedance as compared to the control electrode. Therefore, the low frequencies are attenuated because of the increased impedance of the condenser I5 at low frequencies. In Fig. 2, however, the parallel combination of resistance I! and condenser I5 are effectively in series with the relatively high impedance of the control electrode 5 so that no appreciable attenuation of low frequencies results.

As previously explained, the diiference in fixed potential across the storage condenser I5 due to that which is applied through potentiometer tap 2| does not affect the rate of change of voltage across the condenser I5 because the potential derived from the anode supply source is a static potential to which any further charge on the condenser tendsto approach. 7 r

Fig. 3 shows another modification of this invention in which the synchronizing impulses are inserted in the cathode circuit across resistor 21. It will also be noticed that the winding l3 of the transformer}! is returned to the potentiometer tap 2| rather than to ground, as shown in Fig. 1.

Fig. 4 shows still another modification of the invention. The synchronizing voltage impressed in series with the control electrode circuit across resistor 25 while the transformer winding I 3 is returned to a position between resistances l1 and 25.

It will be noticed that the synchronizing pulses may be fed to the oscillation generator at a number of difierent positions because it is onlynecessary to disturb the control electrode potential by an amount sufficient to cause it to pass the cut-off value of the tube and thus to start another cycle. The point at which the synchronizing pulses are to be fed can then be determined by the. synch. separating circuit which may be such that it will feed into one value of impedance better than another.

While a number of systems for carrying this invention into effect have been indicated and described, it will be apparent to one skilled in the art thatthis invention is by no means limited to the particular organization shown and described but that many modifications may be made without departing from the scope of this invention as set forth in the appended claims.

I claim as my invention:

1 In an oscillation generator, the combination of an electron discharge device provided with a cathode, an anode and a control electrode, a circuit connected between said anode and cathode, means for coupling said circuit to said control electrode to cause said control electrode to go more positive in response to an increase in anode current and more negative in response to a decrease in anode current, said coupling means including a condenser to receive a charge during a relatively small part of each complete operating cycle for applying to the control electrode a potential more negative than that required for cut-off, a voltage source, means for applying from said source a voltage between said anode and said cathode, a potential divider connected across said voltage source and having an intermediate tap connected to said control electrode through a resistive element, said tap being located at a point on said divider where the voltage is such that the variations in bias on the control electrode resulting from any variations in the voltage from said voltage source cause the frequency of oscillation .of said generator to remain substantially constant.

2. In an oscillation generator, the combination of an electron discharge device provided with a cathode, an anode and a control electrode, a

condenser to receive a charge during a relatively small part of 'each complete operating cycle for applying to the control electrode a potential more negative than'that required for cut-off, a transformer'having a winding in a series circuit including said anode, cathode and a potential divider, said transformer having another winding in a second series circuit including said control electrode, cathode, condenser, and a portion of said potential divider.

3. In an oscillation generator, the combination of an electron discharge device provided with a cathode, an anode and a control electrode, a series circuit connected between said anode and cathode and including a source of anode potential, a potential divider across said source, a series circuit connected between said control electrode and said cathode and including a condenser to receive a charge during a relatively small part of each complete operating cycle for applying to said control electrode a potential more negative than that required for cut-off, a resistive element, and a portion of said potential divider and means for coupling said circuits together in such a manner to cause said control electrode to go more positive with respect to said cathode in response to an increase in anode current and'more negative in response to a decrease in anode current.

4. An oscillator for producing electrical impulses periodically, said oscillator comprising an electron discharge tube having acontrol grid and having a. plate circuit and a grid circuit, said circuits being inductively coupled by means of a transformer having a primary winding in said platecircuit and a secondary winding in said grid circuit, said coupling being in such direction that said control electrode is made more positive in response to an increase in plate current, a grid condenser connected in series with said secondary winding whereby said condenser receives a charge of'the proper polarity and of sufiicient -magnitude to block said tube, and a gridresistor for causing said charge to leak 01?, the

values of said grid condenser and said grid resistor being so high and the damping of said plate and grid circuits being so great that each cycle of operation of the oscillator includes a damped sine wave followed by a period of rest, said damping being so great that said grid is driven positively only once during the occurrence of said damped wave, 'a'direct-c'urrent voltage source in said plate circuit for supplying a positive operat- :such that said undesirable voltage variations have substantially no effect upon the periodic rate of the impulses produced by said oscillator.

ALDA v. BEDFQRD.