Generally, production of a sawtooth wave involves the charge and discharge of a capacitor through resistors which differ greatly in size between the charge and discharge circuits.
An introduction to this concept has been covered under the heading of R-C Circuits. It has been shown that, to produce a sawtooth waveform, we need a simple circuit consisting of a source of voltage, a single pole double throw switch, resistors and a capacitor.
The capacitor is charged through a high value of series resistance. The voltage across the capacitor at any instant of time has been shown in Figure 43. It will be noted that the initial portion of this voltage -time curve is essentially a straight line. If we can short-circuit the capacitor before extreme curvature of the charge-time has occurred, and immediately initiate another charging cycle, we have produced a sawtooth wave . This sequence of events could be accomplished by the use of a mechanically operated switch. (See Figure 51.) Since this entire operation must occur in a few millionths of a second, it is obvious that the use of such a mechanically operated switch is impractical. For this reason we will resort to some of the properties of electron tubes to accomplish the switching sequence.
While modern television receivers employ vacuum tube oscillators and vacuum tube wave shaping circuits to accomplish the production of the ideal sawtooth scanning motion previously described, it will be instructive to examine the earlier forms of circuits employ ed. These methods are no longer used in television sets, but they are undoubtedly familiar since their application is still common in the cathode-ray oscilloscope.
NEON TUBE RELAXATION OSCILLATORS: The familiar neon gas-filled tube, employed in sign lighting, is one of the simplest automatic switches for the purpose of short-circuiting the capacitor at the proper instant to produce a sawtooth voltage wave. A gas-filled tube having a pair of electrodes and connected to a source of electrical potential exhibits interesting properties as the voltage across the electrodes is gradually increased. No electrical current will flow through such a tube until the voltage reaches a value known as the "ionization potential”. The tube, until this voltage has been reached, has acted as an open circuit or as an extremely high resistance. However, when the ionization potential has been reached, the pressure, or voltage, tears electrons from the atoms of the gas and leaves such stripped atoms with a positive potential. Under this set of conditions the free negative electrons are rapidly collected by the positive electrode, the positively charged atoms are correspondingly attracted to the negative electrode and current passes through the tube. The resistance of the tube suddenly changes and it can be considered as a voltage operated switch. Figure 52 shows the method of using such a gas-discharge valve as a switch across a capacitor.
Fig. 52. Neon Tube Relaxation Oscillator
It will be seen that when the charge in the capacitor has produced a voltage across the tube equal to its ionization potential, the tube will suddenly conduct and start to discharge the capacitor. An interesting property of the tube, is the fact that once conduction of current has started it will continue even though the voltage has dropped below the original ignition point. Current will continue to flow until a lower level of voltage, known as the de-ionization potential, is reached. At this point the atoms of the gas regain their free electrons and the tube returns to its non-conducting condition.
The charging cycle from the voltage source through the series resistor is resumed and the cycle continues until ionization again occurs.
This sequence of events is shown diagrammatically in Figure 52. Such an automatically operated switching circuit is known as a relaxation oscillator because during the periods between its trip-over action the circuit is under normal (or relaxed) charging condition.
THYRATRON RELAXATION OSCILLATORS:
The negative grid-controlled thyratron is an improved form of gas-discharge tube. Such a tube is essentially the same in action as the simple neon lamp previously described, with the following exceptions:
1. A source of electrons in the form of an electrically heated cathode, supplies the electrical current for the discharge portion of the cycle.
2. The trip-over, or triggering action, is under the control of an additional element similar to the grid of the familiar radio vacuum tube. This element is normally held at a negative potential and prevents current conduction between the cathode and plate, or anode, by its repelling action on the electrons emitted by the cathode.
3. The gas normally employed is not neon but preferably mercury or argon.
Fig. 53. Thyratron Relaxation Oscillator
A relaxation oscillator can be built with this tube and its action is more easily control led than that of a simple neon gas tube. Figure 53 shows the basic circuit of a thyratron saw tooth generator. In order to assure that the rise of the charging potential is linear, only a small portion of the plate or "B" voltage is allowed to charge the capacitor. The grid is held at a sufficiently negative potential to in sure that there is no plate current. In the diagram this grid potential is provided by a bias battery. Trip-over of the circuit can be produced by a positive pulse of voltage applied to the grid. Once initiated, plate current will flow until the plate voltage has dropped to a point which corresponds to the de-ionization potential as described in connection with the neon oscillator.
A circuit of this type is frequently employed in cathode-ray oscilloscopes and has also been used in pre -war television sets both in this country and abroad. Post-war receivers no longer employ gas tube relaxation oscillators due mainly to the fact that such oscillators are not sufficiently reliable in operation with fluctuating power line voltage, temperature and time.
VACUUM TUBE SAWTOOTH GENERATORS: We have seen that the requirements of a de vice to produce a sawtooth voltage waveform are basically those of applying a voltage to a capacitor through a series resistance and after the capacitor has reached a pre-determined charge, removing the voltage by a virtual short circuit. This order of events can readily be accomplished by high vacuum type tubes rather than gas tubes.
In general, modern television sets have employed three distinct types of circuit arrangements, or combinations of these circuits, to produce sawtooth waveforms.
1. The multivibrator. This circuit arrangement admits of many variations, the most popular of which is the cathode-coupled version.
2. The blocking oscillator. This type of circuit permits the formation of a short-time interval pulse of energy which can be used to produce the sawtooth wave across a capacitor directly associated with the oscillator tube, or can employ the pulse to trigger a "discharge" tube which acts as a switch across the capacitor.
3. The sine-wave oscillator. An oscillator of the correct frequency supplies the timing voltage for the discharge tube and has sine-wave output modified to the form of short time pulses by intermediate or wave-shaping circuits. These pulses are then used to operate a discharge tube to produce sawtooth waves.
THE MULTIVIBRATOR CIRCUIT . One of the most popular of television scanning generator circuits is the multi vibrator. This is an o the r form of relaxation oscillator which employs vacuum tubes, resistors and capacitors in a feedback arrangement.
The multivibrator derives its usefulness from the fact that tubes can be made to act as automatic switches to control the charge and discharge of capacitors. This produces a sustained output of rectangular wave form, the frequency of which can easily be controlled by the horizontal or vertical synchronizing pulses.
Several versions of the multivibrator circuit are found in modern television sets but since they are derived from a basic or conventional type we will first examine the operation of the fundamental circuit.
The Conventional Multivibrator
The basic free-running multivibrator might be considered as a two stage resistance-capacitance coupled amplifier with over all feedback applied by means of a capacitor connected from the output of the second stage to the input of the first.
Figure 54A shows a familiar two stage audio amplifier. The addition of one circuit element ( C1 shown in dotted lines) converts this amplifier into a free running multivibrator type oscillator. Figure 54B shows a symmetrical rearrangement of the same circuit as it is usually presented in text books and receiver schematic diagrams.
Since, in a single stage resistance coup led amplifier, the phase of the plate voltage is inverted (180 degr.) with respect to that of the input or grid voltage, it follows that the output volt age of the second stage (T2) of the two stage amplifier will have again been inverted and will be in phase with the input (T1) voltage. Capacitor C1 of Figure 54 will therefore impress upon the first grid, a voltage of the proper polarity to increase or augment the original input voltage and oscillation can take place.
To follow through the manner in which oscillation starts and is maintained in this circuit, let us assume that the cathodes are heated and that B+ voltage is suddenly applied. Both grid circuits are returned to their respective cathodes through grid resistors and at the instant of applying B+ voltage the grids will be substantially at cathode potential or zero bias.
Grid current as well as plate current will start to flow in each tube.
(A) AS A TWO STAGE AMPLIFIER WITH CAPACITIVE OVERALL FEEDBACK. (B) SAME CIRCUIT AS USUALLY DRAWN.
Fig. 54. The Conventional Multivibrator.
It will be instructive at this point to list in numerical order with reference to Figures 54 through 57, the sequence of events which produces the sustained rectangular shaped output wave of the multivibrator.
1. Since the resistance of the internal cathode to grid path, under this initial condition of zero bias and high grid current , is much lower than the resistance of the grid resistors R1 and R2, the two capacitors C1 and C2 will begin charging from the B+ supply through resistors R4 and R3 respectively. This charging path is shown by the arrows of Figures 55A and 55C.
2. Now, if the characteristics of both tubes and the value of the circuit elements were exactly matched, the charging rate of both capacitors would be identical and plate current rise of both tubes would occur simultaneously. Under these conditions, a state of equilibrium would be reached and the circuit would not produce oscillation. Such a set of conditions is not met in practice and a balance is not established.
3. In the practical case, one of the tubes will start to conduct plate current sooner than the other. This could be due to a number of causes such as, lower plate resistance, hotter cathode or slightly lower plate load resistor.
Let us assume that the plate current of T1 has started to rise a fraction of a second ahead of that of T2.
4. This rise of plate current will be accompanied by a drop in plate to cathode resistance and a corresponding drop in plate to cathode voltage. Figure 56A shows this set of operating conditions. The low plate resistance of T1 forms a discharge path for C2 as shown in Figure 55D.
Fig. 55. Capacitor Charge and Discharge Paths in a Multivibrator
Fig. 56. Tube Operating Conditions in Multivibrator Oscillators A
5. The discharge current of C2, flowing through the high value of grid resistor R2, develops a high negative grid bias on the grid of T2. This drives the tube beyond plate current cut-off as shown in Figure 56B. The bias developed by this discharge action may be as high as 30 to 50 volts in the example shown.
6. Since the plate current of T2 has been cut off, its plate to cathode voltage becomes that of the B supply (see Figure 56B), and will remain at that value until the grid voltage has arrived at such a point on the discharge -time curve that the grid is no longer cut off.
Note: Since the above set of conditions has brought the cycle of operation to one of the two stable or "relaxed" operating points of the circuit it would be helpful to summarize the changes of circuit voltages which have occurred over the period covered by steps 1-6.
T1 -- Plate to cathode voltage at its minimum value and steady. Tube conducting.
T2 -- Plate to cathode voltage at its maximum value and steady. Tube not conducting.
T1 -- Control grid voltage zero and steady.
T2 -- Control grid voltage highly negative but falling exponentially with time as C2 discharges through R2.
7. The time required for C2 to discharge will depend on the time constant of the discharge circuit which comprises C2, R2 and the plate resistance of tube T1. See Figure 55D. The negative voltage across R2, which constitutes the grid bias of T2, finally decreases to a value which will allow T2 to conduct heavily.
Figure 57 shows the wave forms of the grid and plate voltages of both tube s as a function of time. The part of these waves between (a) and (b) in this figure cover the steps which have been outlined up to this point.
8. As T2 starts to conduct, conditions in this tube become identical to those described in step 4 for T1, except that the tubes and capacitors have exchanged functions, and the discharge path of capacitor C1 is now as shown in Figure 55B.
9. The discharge current of C1 flowing through R1 now biases T1 beyond cut-off as de scribed in step 5.
Fig. 57. Typical Waveforms of a Symmetrical Multivibrator Showing Square Wave
Switch Action.
10. Since the plate current of T1 has been cut off, the plate to cathode voltage assumes the value of the B+ supply in similar fashion to T2 in step 6.
11. The rise in plate voltage of T1 is impressed on capacitor C2 starting a charging cycle as shown in Figure 55A. 12. Since the internal grid to cathode path of T2 is conductive due to the zero grid bias condition, the charging resistance is very small and C2 is charged very rapidly. This is shown at time (b} of Figure 57.
A new cycle has been started and we have arrived at a similar set of conditions to those described at the end of step 6.
We can now summarize the conditions of circuit voltage and compare them with those found at the end of step 6.
T1 -- Plate to cathode voltage at its maximum value and steady. Tube not conducting.
T2 -- Plate to cathode voltage at its minimum value and steady. Tube con conducting.
T1 -- Control grid-voltage highly negative but falling exponentially with time as C1 discharges through R1.
T2 -- Control grid voltage zero and steady.
Fig. 58. Waveforms of an Asymmetrical or Unbalanced Multivibrator
It will be noted that the new set of conditions which represent the o the r stable or relaxed operating point is the same as before except that the tubes and grid circuits have changed places with one another. This cycle of events is shown in the waveform diagrams of Figure 57 between times (b} and (c). This generation of a square wave will continue at a frequency that is determined by the charge and discharge time constants of the coupling net works R1-C1 and R2-C2.
In this symmetrical circuit it has been assumed, but not previously stated, that the corresponding grid resistors, plate resistors and coupling capacitors are equal. When this is the case, the time constants are equal and the output wave shape from the plates is essentially identical. The frequency of this multi vibrator can be changed by altering either the resistors R1 and R2 or the capacitors C1 and C2. A decrease in value of the time constant will increase the frequency. If the R values or the C values are changed equally the output wave will remain symmetrical.
The Asymmetrical or Unbalanced Multivibrator
In order to use the multivibrator to produce the type of sawtooth wave required for television scanning, it is necessary that succeeding square waves be unequal in length or spacing, and for this reason the time constant of the R-C circuit of one tube is deliberately made much greater than the time constant of the other. A multivibrator of this type is called asymmetrical. Figure 58 shows the waveforms obtained when the circuit constants of the symmetrical multivibrator just described, are changed in such a manner that the product (R1xC1) in the grid circuit of T1, is much smaller than the product (R2xC2) of T2.
Waveform D of Figure 58 shows a short time pulse of plate current occurring in tube T2 once each cycle, and it is this pulse which we will employ to produce the scanning sawtooth in proper time relationship to the scanning of the camera tube at the transmitter.
Use of the Multivibrator to Produce Sawtooth Scanning. Figure 59 shows a circuit similar to those which we have discussed under the headings of symmetrical multivibrators and also as an asymmetrical multivibrator. By the addition of two new circuit elements we can convert this arrangement into a method of generating sawtooth voltage waves for the control of the electron beam by either the horizontal or the vertical deflection plates of an electro static television picture tube. These new circuit elements are C3 and C4 (Figure 59). C3 is used as a coupling capacitor to connect the multivibrator circuit to a source of synchronizing pulses which are part of the transmitted television signal. The function of the synchronizing pulses and their action in the control of the frequency of the multivibrator will be covered in a later section.
At this time, the additional circuit element which concerns us, is the capacitor C4 connected between the plate and the cathode of T2.
Fig. 59. Sawtooth Scanning Wave from Asymmetrical Multivibrator
For horizontal line scanning frequency of 15,750 cycles per second, the circuit is ~o arranged that the time constant (R1xC1) 1s approximately 1/9th of .the time constant (R2xC2). Under these conditions the plate current of T2 will consist of short pulses as shown in Figure 58. These pulses represent a condition of low resistance, and it is obvious
that during the time shown as the "conducting period", tube T2 will act as a virtual short circuit across capacitor C4. In this manner the multivibrator is acting as a periodic switch and has fulfilled the requirements for the production of a sawtooth wave as covered in pages 37 and 38.
This type of controlled multivibrator has been used for both horizontal and vertical deflection scanning, in many of the lower priced electrostatically deflected television sets.
The voltage of the sawtooth wave across capacitor C4 (Figure 59) is of insufficient magnitude to produce the required deflection.
Amplifiers are therefore needed, and usually take the form of push-pull circuits incorporating a phase inverter . Another significant difference between the circuit Figure 59 and Figure 54 is the fact that R2 has been made variable. This variable resistor is one of the major controls of a television receiver. From the previous discussion of multivibrator theory it is obvious that variation of R2 will alter the length of the portion of the operating cycle shown controlled by (R2-C2) in Figure 58.
This represents the portion of the sawtooth wave which is active in scanning the face of the cathode ray picture tube during video modulation. Since this variable adjustment permits the multivibrator to be "locked-in" with the synchronizing pulse and "held" in step by that pulse, it has been known in engineering circles as a "hold" control. This nomenclature has been carried over into the trade and many modern television sets label this control knob as "horizontal hold" or as "vertical hold" de pending upon the particular function concerned.
CATHODE-COUPLED MULTIVIBRATOR: An interesting variation of the multivibrator which is becoming practically a standard for inexpensive electrostatically deflected type television receivers is the cathode-coupled multivibrator or "Potter" circuit. This circuit is shown in Figure 60. A significant difference between this circuit and the conventional multivibrator is the fact that feedback is accomplished in two ways. The coupling capacitor C2 serves, as before, to transfer charges from the plate of T1 to the grid of T2. In addition, this circuit employs a cathode bias resistor which is common to T1 and T2. This common cathode resistor is responsible for the unique action and the name of the circuit.
The second tube (T2) functions as a switch or discharge tube for capacitor C4 which produces the sawtooth waveform. From the theory of ...
Fig. 60. Cathode -Coupled Multivibrator as Sawtooth Wave Generator
..... the conventional and asymmetrical multivibrators as previously discussed, the action of this cathode-coupled version can be readily understood. As before, we will assume that the cathodes of the tubes are heated and that B+ potential is suddenly applied. Let us again follow through in numerical order the sequence which allows this circuit to generate asymmetrical pulses.
1. Capacitor C2 will charge through the path consisting of R3 and the grid to ground or B- circuits of T2. This action will occur very rapidly since the grid of T2 is initially at zero potential.
2. Plate current will start to flow in both T1 and T2 causing a bias voltage for both of these tubes to be developed across the cathode resistor, Rk.
3. This bias voltage will immediately start to decrease the plate current of both tubes which were initially in a conductive condition since the control grids were at substantially zero potential.
4. The flow of plate current through T1 causes a lower plate to cathode voltage drop and a correspondingly lower plate resistance of this tube.
5. The low resistance path of T1 initiates the discharge of coupling capacitor C2 through R2. As in the conventional multivibrator previously discussed, this current flow through R2 produced a high negative bias on the control grid of T2 which immediately sweeps the tube to beyond its plate current cut-off point. It will be noted that this circuit differs from the conventional multivibrator in that there is no coupling capacitor between the plate of 'P and the grid of T1. For this reason, T2 is immediately thrown into a condition of plate current cut-off.
6. As in our discussion of the conventional multivibrator, it will be constructive at this time to summarize the voltage conditions which have occurred up to this point: T1--Plate to cathode voltage at its minimum value. Tube conducting.
T2--Plate to cathode voltage rising along a portion of the charging curve which is substantially linear. This rise of plate voltage is charging capacitor C4 and initiating the first part of what will eventually become a sawtooth wave of voltage.
T1--Control grid voltage negative and steady. The tube is self-biased by its own plate current flowing through the common cathode resistor Rk.
T2--Control grid potential highly negative and exponentially diminishing in value.
Fig. 61. Voltage Waveforms in Cathode-Coup led Multivibrator
7. T2 has been cut off during this period and no plate current has been flowing. C2 has been discharging through the path R2, Rk and the cathode to plate circuit of T1. It is interesting to note, that as the plate current of T2 was cut off by the high negative bias produced across R2 by the discharge of C2, the plate to cathode voltage did not immediately assume the value of the B+ supply since the plate voltage of this tube was maintained by the charge existing in C4, which started simultaneously with the closure of the B+ circuit. For this reason, the plate voltage wave of T2 instead of being rectangular form will be of sawtooth shape due to the charge flowing into C4 (if C4 is removed from the circuit then the plate to cathode volt age of T2 would rise immediately to the B+ value since the grid of this tube is cut off by the high negative voltage resulting from the discharge of C2).
8. As in the other types of multivibrators already discussed, when the bias of T2 falls to a value equal to the cut-off grid potential, T2 will start to conduct.
9. When conduction occurs in T2, C4 will be discharged rapidly through the plate to cathode circuit of this tube and its plate to cathode voltage will drop to its minimum value.
See Figure 61. The sequence of events to this point has resulted in the production of a saw tooth wave of voltage across capacitor C4.
Thus far, the action of the circuit has resembled that of the conventional and asymmetrical multivibrators previously discussed.
10. A significantly different action now takes place. The sudden pulse of plate current which occurs when T2 conducts, flows through cathode resistor Rk, and since this resistor is common to the cathode circuits of both T1 and T2, the voltage produced by this plate current pulse immediately drives the grid of T1 negative with respect to its own cathode.
11. This negative bias causes a sudden increase in the plate to cathode resistance and the plate to cathode drop of T1. The effect of the sudden increase of T1 plate voltage is to cause C2 to charge, thus instantaneously impressing a positive voltage on the grid of T2, momentarily increasing the value of the plate current pulse of T2 which started the cycle.
12. The cumulative increase of plate current flowing through common cathode resistor Rk finally produces sufficient negative bias on the grid of T1 to completely cut off the plate current, and the plate to cathode voltage rises to its maximum value.
13. Capacitor C2 has become charged and the plate current of T2 "relaxes". 14. This decrease in current flow in Rk reduces the bias of T1. As T1 conducts, C2 is discharged through R2, Rk and the plate circuit of T1. This current flow in R2 drives the grid of T2 to cut-off and the cycle is repeated.
The fact that a sudden and cumulative action was produced in this circuit by the coup ling of tubes T1 and T2 through a common cathode bias resistor, accounts for its designation as a cathode-coupled multivibrator. The reasons for the preference for this type in television design over the asymmetrical multi vibrator previously discussed, are as follows:
1. It can be triggered and controlled by a negative pulse of voltage which often leads to simplification of control circuits.
2. Its sudden and cumulative pulsing action in tube T2 permits a higher ratio of linear sweep time to return time.
3. Simple variable resistors R2 and R4 of tube T2 (see Figure 60) permit control of both the frequency of scanning and its amplitude.
The wave shapes of the voltages at various points in this circuit, as a function of time, are shown in Figure 61. These wave shapes are identified by letters enclosed in 'diamonds' in Figure 60. In all cases voltages are measured to B- or ground.
THE BLOCKING OSCILLATOR. Another type of vacuum tube circuit for the production of controlled sawtooth voltage waves is known as the "blocking oscillator". The blocking oscillator is becoming increasingly popular in modern television sets as a means of producing either vertical or horizontal scanning. This popularity is due to the fact that its reliability of operation and ease of adjustment are de pendent upon circuit elements such as trans formers, resistors and capacitors rather than being dependent upon the characteristic of a vacuum tube which is employed in the circuit.
Vacuum tubes are manufactured to rather wide commercial tolerance values and are subject to change of characteristics with age.
Fig. 62. Blocking Oscillator as a Sawtooth Generator
Figure 62 shows the blocking oscillator in its simplest form. It appears upon casual inspection to be a Hartley oscillator employing an iron core transformer. In essence it is such an oscillator, but instead of providing sustained sine wave oscillations it produces short time pulses of energy with correspondingly long intervals of relaxed action. For this reason, it is classified as another form of relaxation oscillator. Two significant differences distinguish this circuit from the common Hartley oscillator, namely:
a. The time constant of the grid resistor R1 and the grid capacitor C1 is such that long periods of blocked plate current occur between short periods of plate current conduction.
During these short conductive periods oscillation takes place.
b. The natural period of oscillation of the transformer, with its associated distributed and lumped circuit capacitances, is such that the desired pulse time approximates one-half cycle of the frequency at which the circuit would oscillate if it were of the continuous sinusoidal type.
As in the case of the other types of saw tooth oscillators it will be instructive to follow through, in sequence, the various actions which take place in this circuit.
1. We will again assume that the cathode of the tube is heated and that the plate circuit is suddenly closed to provide B+ potential.
Since the grid is initially at cathode or zero potential, plate current will start to flow through the primary of the transformer. This sudden rush of current will set up a magnetic field in the core of the transformer, inducing a secondary voltage across the grid winding.
The direction of these windings is such that the primary current will cause a positive potential to appear at the grid with respect to cathode or ground.
2. The positive voltage thus applied to the coupling capacitor C1 causes the grid to be come more positive than the cathode. The grid then attracts electrons from the emitted cathode current, starting grid current through resistor Rt.
3. Simultaneously, the increasingly positive grid potential causes the plate to draw still more current and the action is cumulative until plate current saturation is reached. As the plate current reaches a steady maximum value no further change of current occurs through the primary winding of the transformer.
4. Induction of voltage into the secondary depends upon the change of magnetic flux and therefore at this time the secondary voltage of the transformer ceases to rise.
5. As the grid tends to become less positive ( C1 discharging through R1), the plate current through the primary begins to decay and the magnetic field linking the secondary coil starts to collapse. Note: The time taken for this sudden rise and decay of grid voltage is governed by the natural resonant frequency of the transformer with its associated circuit capacitances.
6. The collapsing field in the transformer due to the dropping plate current induces a secondary voltage which is in the opposite direction to the original plate current pulse. This causes C1 to discharge through resistor R1 which drives the grid more and more negative, hastening the decay of plate current and finally causing it to reach a cut-off point. While this action of reversal of grid voltage and cut-off of plate current has taken considerable time to describe, in reality it is practically instantaneous.
7. From this point, action in the tube follows that described in the multivibrator and the grid potential follows an exponential curve of R-C discharge until the point of plate conduction is again reached.
Fig. 63. Voltage· Waveforms in Blocking Oscillator
8. The time taken for this discharge of C1 is dependent upon the time constant (R1+R2) x C1.
9. As the tube starts to conduct again, oscillation starts and the cycle is repeated. It will be seen from the curve of Figure 63 that the plate voltage of the tube is nearly steady and at its maximum value between the occurrence of these oscillatory pulses. It is evident that we have fulfilled the conditions of sawtooth charge and discharge of capacitor C3 in Figure 62 and have therefore produced a sawtooth scanning wave. As in previous circuits, the Grid resistor R1 can be made variable as a frequency control ( "hold" control) and a plate circuit resistor made variable as a width or height control.
Continuing the study of the blocking oscillator, it is important to note that in the circuit of Figure 62, the sawtooth generating capacitor C3, which is connected from plate to cathode of the tube, modified the shape of the plate voltage wave. We have seen that a similar type of action occurred in the cathode-coupled multi vibrator circuit of Figure 60. If, in this case, capacitor C3 had not been present the voltage wave occurring between plate and cathode of the tube would have the shape of Figure 63B. The excursion of the wave above the B+ axis at point x is caused by the stored energy in the primary of the transformer. This action is analogous to that of the constant current Heising modulation choke of a transmitter. We have already discussed a similar phenomenon in connection with the "flyback" system of developing high voltage on pages 29 and 30. When the capacitor C3 is connected the voltage wave from plate to cathode assumes the shape shown in Figure 63C. This is the desired sawtooth deflection wave except for the distorted section shown at y. This sudden rise of the curve at this point is due to the additional contribution of charging voltage which occurred when the magnetic flux of the transformer collapsed as explained above. The plate voltage was prevented from following the curve of Figure 63B by the terminal voltage contribution of capacitor C3. In the practical use of this circuit in television receivers the distorted section of the wave at y is of no con sequence since it is "blanked out" as will be explained later.
Fig. 64. Blocking Oscillator and Discharge Tube as a Sawtooth Generator
Discharge or Trigger Tube . In many modern television sets, pulse generators are not used as the direct control of the sawtooth charge and discharge of the capacitor, but rather are used to trigger an additional tube known as a "discharge tube". This tube in turn short-
circuits or discharges the capacitor. Figure 64 shows a blocking oscillator, of the form previously described, whose grid is connected directly to a second tube the only function of which is to conduct plate current, and in consequence discharge capacitor C2 at the proper instant to produce the sawtooth wave. The introduction of this extra tube provides independence of functions as compared with the simpler form of Figure 62 which might suffer from some interaction of controls. The discharge tube can be used with any type of relaxation oscillator in which the plate current occurs over short intervals of the operating cycle.
SINE WAVE GENERATORS. It is possible to employ the familiar sine wave oscillator to produce pulses which, in turn, will trigger a discharge tube associated with a sawtooth wave-forming capacitor. Only a short portion of the sine wave is used and it is passed through what are known as "clipping" stages to "bite" off a small section of the wave. In this manner the output of the clipper is in the nature of a pulse. The utility of this type of circuit will be discussed later when we consider the control of scanning by the synchronizing pulses.
These oscillators are biased to run in a similar fashion to class C transmitter technique in which plate current is cut off for part of the cycle. Part of the necessary clipping of the sine wave to produce a pulse is already accomplished in the oscillator. Their action will be covered in greater detail when we discuss actual television deflection circuits.
SUMMATION OF THE THEORY OF MULTIVIBRATOR AND BLOCKING OSCILLATOR ACTION:
1. Multivibrator circuits constitute a means of employing vacuum tubes as electronic switches to control the charge and discharge cycles of a capacitor for the production of sawtooth waves.
2. This switching action by the tubes is practically instantaneous and is limited in timing by the circuit elements, rather than by the tube itself. It is interesting to note that the velocity of the electrons in a tube, which is suddenly driven to saturation after a condition of plate current cut-off, can approach one-tenth of the velocity of light or approximately 19,000 miles per second.
3. When an electron tube acts as a voltage operated switch, it is suddenly changed from a condition of "no plate current" to one in which plate current is limited only by the ability of the cathode to supply electrons (plate current saturation). These two states are illustrated in Figure 56. Since this action is used in other sections of a television receiver. it is valuable to become thoroughly familiar with the steps involved.
4. Multivibrator circuits may be either symmetrical or asymmetrical. In the first case, the result is the production of square voltage waves. These resemble a '' Roman Key" pattern. The symmetrical multivibrator is useful at the television transmitter in the production of the complex television signal. It has no present use in the receiver and for that reason we are interested only in the asymmetrical type.
5. The asymmetrical multivibrator produces voltage output waves in which a short pulse of rectangular shape is followed by a long space or "gap". This short pulse of voltage can be used to switch a vacuum tube from a condition of plate current cut-off to a condition of plate current saturation.
6. If a capacitor of the correct value has been connected across the plate of this tube, the long periods between pulses will have been occupied by the gradual or exponential charging cycle, and the voltage across the capacitor will rise in a substantially linear fashion.
7. When the short rectangular pulse of voltage is suddenly applied to the grid of the tube, and the tube becomes conductive, it will short circuit the capacitor and start a new sawtooth wave.
8. An important reason for the use of this means of generating sawtooth waves , is the fact that dual triode tubes such as the 6SN7 can be employed in lieu of the two tubes shown in the illustrations.
Dual tubes in one envelope present a difficult control problem in the plant of any tube manufacturer. For this reason, erratic scanning action in a television set can often be remedied by tube interchange. Fortunately, there are many other circuits, such as clippers, sync separators, etc., where the same tube type is used, and where its characteristics are not critical. The cure for unstable multivibrator action is often found in the simple interchange of tubes.
9. Checking the values of resistors and capacitors, determining that there are no open or short circuits, and substituting tubes, constitutes the normal service procedure in multi vibrator scanning generators.
10. The blocking oscillator requires only one vacuum tube, as contrasted with the multi vibrator which depends upon a phase rotation of 360° through two tubes. The phase of feed back in the blocking oscillator is provided by the relationship of transformer windings.
11. The characteristics of the transformer determine the length of the conduction time or "closed switch" part of the cycle. This time is approximately one half cycle of the transformer resonant frequency as tuned by its associated capacitances.
12. The "relaxed" time between "switch on" or conduction periods is determined by the time constant of the grid circuit capacitor and its discharge resistors. The grid resistor can be made variable to control the frequency of oscillation, and allow synchronization with the transmitted signal.
13. In the blocking oscillator, plate to grid feedback through the transformer sets up a grid current which charges the grid capacitor, causing the grid to be driven positive momentarily. This, in turn, causes the plate current to rise to its saturation value. The plate current then begins to decay, and reversal of direction of the induced grid voltage causes the grid capacitor to discharge through the grid resistor. The grid is then driven very negative cutting off plate current. The cycle repeats itself as soon as the voltage across the resistor reaches a value which will allow the grid to again initiate plate current.
14. Discharge or "trigger" tubes are frequently used with pulse generators, such as the blocking oscillator, to provide separate control of scanning wave size and shape. An other function of discharge tubes is to provide a means of producing special wave shapes for particular scanning requirements. This subject will be covered in detail later in the text.