|Home | Articles | Forum | Glossary | Books|
Like their triode vacuum-tube counterparts, transistors can be employed to detect, or demodulate, a modulated RF carrier signal and thereby develop an audio voltage. The first circuit discussed in this Section is for a PNP transistorized detector.
Crystal diodes are also employed as detectors in many transistor receivers. The second circuit discussed in this Section is for an IF amplifier and diode detector, with AGC applied to the IF amplifier.
PNP DETECTOR CIRCUIT
Figs. 1 and 2 show two successive audio half-cycles in the operation of a transistor detector. The necessary components of this circuit include:
R2-Voltage-divider and base-bias resistor.
R3-Emitter stabilizing resistor.
R4-Collector load resistor.
C1-IF tank capacitor.
C2-IF filter capacitor.
C3-Emitter bypass capacitor.
C4-IF filter capacitor across load resistor R4.
C5-Output coupling and blocking capacitor.
T1-IF tank transformer.
M1-Battery or other power supply.
Identification of Currents
The following separate and distinct electron currents are at work in this circuit. The actions occurring in this or any circuit cannot be understood until the movements of these cur rents are clearly understood.
1. IF tank current (solid blue).
2. Base-emitter biasing current, a direct current which pulsates at the intermediate frequency (solid green).
3. Voltage-divider current (dotted green).
4. Two intermediate-frequency filter currents (dotted blue).
5. Collector-emitter current (solid red).
6. Audio filter current (dotted red).
Details of Operation
The intermediate-frequency input current (in solid blue) oscillates in the tank circuit consisting of C1 and the primary of transformer T1. This IF tank current induces a similar current in the secondary winding. With this secondary current there will always be associated the so-called secondary voltage, also known as "back electromotive-force" (back emf) . Once during every cycle of this intermediate-frequency current, the upper terminal of the secondary winding will be driven to a peak of negative voltage. When this happens, the transistor will conduct the maximum base-emitter biasing current because, in the PNP transistor, electrons invariably flow from the base to the emitter against the direction of the emitter arrow). Hence, a more negative base voltage will naturally increase this flow.
Between each negative peak of induced voltage there will be an equal-sized peak of positive voltage. On these positive peaks, the transistor will conduct the minimum base-emitter current.
Since this current always flows in the same direction through the transistor, the voltage alternations which were induced across the secondary winding of T1 are converted into current pulsations through the transistor.
The complete path of this base-emitter biasing current, which is shown in solid green and which pulsates at the intermediate frequency, begins at the negative terminal of battery M1. It flows to the left and upward through resistor R2 and the secondary winding of T1. Then it heads to the right into the base of the transistor and out the emitter and downward through resistor R3, to the common ground connection, where it has free access to re-enter the positive terminal of the battery.
In addition to the pulsations which occur at the basic inter mediate frequency, this base-emitter current also varies in amount according to the modulation carried by the carrier signal.
Fig. 3A shows a fairly conventional graphical representation of the so-called modulated carrier signal, or waveform. Many hundreds of even thousands of the carrier cycles will occur during one audio cycle, and their strength will periodically rise and fall in accordance with the modulation imposed on the carrier at the transmitter. These variations in carrier strength make up what is commonly known as the "modulation envelope." As is customary in waveform diagrams of this type, the modulation envelope has been indicated on Fig. 3A. This so-called "modulation envelope" in such diagrams is nothing more than a convenient graphical device which relates the relative strength of individual cycles of the carrier signal and, of course, points up the fact that the strength of these carrier cycles varies in accordance with the audio modulation imposed on the carrier at the transmitter.
Each pulsation of base-emitter current causes one cycle of IF filter current to flow through filter capacitor C2 to ground.
During the negative peak portions of the secondary induced volt age across T1, maximum base-emitter current flows through the transistor X1. Inevitably some electron current is drawn from the upper plate of capacitor C2, and in turn the same quantity of electrons is drawn onto the lower plate. On the positive peak portion of secondary induced voltage, the base-emitter current is reduced to minimum. During these half-cycles of IF, the filter current flowing on either side of capacitor C2 heads downward again to ground.
The amount of base-emitter biasing current which flows through any transistor is determined by the two important biasing volt ages, (at the base and emitter) and, of course, the difference be tween them. The voltage at the base of transistor X1 is determined primarily by the flow of voltage-divider current (in dotted green) upward through battery M1, to the left and upward through resistor R2, and downward through R1 to the common ground. From here, it has easy access back to the positive terminal of battery M1. The resulting voltage at the junction of resistors R1 and R2, being less negative than the power-supply voltage, will cause a flow of electron current across the junction between base and emitter. The amount of this so-called biasing current controls, or regulates, the flow of electron current from collector to base, within the transistor.
(Once this current crosses the difficult "reverse" junction from collector to base, it flows fairly easily from base to emitter, and exits from the transistor at the emitter terminal.) This second current through the transistor is usually called merely the collector current, and will be from 25 to 100 times larger than the biasing current (frequently referred to merely as the "base" current). The collector current, shown in solid red, begins at the negative terminal of battery M1. From here it flows upward through resistor R4, then downward through the transistor from collector to emitter, continuing downward through emitter resistor R3 to the common ground, where it can return to the positive terminal of the battery.
(A) Modulated IF carrier. (B) Base and emitter voltages. (C) Collector current.
In flowing downward through R3, the collector-emitter current will develop across R3 a voltage of negative polarity at the upper terminal and positive polarity at the lower terminal. The negative voltage polarity at the upper terminal of R3 becomes the second of the two important biasing voltages of the transistor.
Resistor R1 and capacitor C2 together constitute a "long time constant" filter-one whose time constant is longer than the duration of a single intermediate-frequency cycle. Consequently, a negative voltage will appear on the upper plate of capacitor C2. This negative voltage, which remains unchanged between the individual IF cycles, has been indicated by the blue minus signs on C2. (A negative voltage stored on a capacitor plate can be most conveniently represented as a group or pool of electrons, the symbol for which is one or more minus signs.)
In Fig. 2, which represents a "Peak" half-cycle of audio voltage, the single minus sign on the upper plate of C2 indicates the negative base voltage which would correspond to a modulation peak. Fig. 3A tells us that during a modulation peak, the individual IF cycles have their maximum value, or strength. Each such cycle drives the transistor base to 3 negative voltage of such value that maximum base-emitter biasing current is per mitted to flow through the transistor. This extra quantity of base emitter biasing current must flow through resistor R2. In so doing, it lowers the negative voltage at the upper terminal of R2, because of the increased voltage drop across the resistor.
The reduced negative voltage at the junction of R1 and R2 is also the voltage applied to the base. Thus, during the modulation peak of Fig. 2 a smaller negative biasing voltage is applied to the base of the transistor. As a result, the base-emitter biasing current which flows continuously through the transistor is reduced.
During the modulation trough of Fig. 1, the individual IF cycles reach smaller peak values and thereby reduce the base emitter current during each cycle. Since this current must flow through resistor R2, the decreased voltage drop across R2 makes the voltage at the junction of R1 and R2 more negative. This is one of the two transistor biasing voltages, and it causes a general increase in the continuous flow of base-emitter current through out the entire cycle.
The voltage which appears on the upper plate of capacitor C2 marks the first appearance of an audio voltage when a carrier signal is being modulated. It varies between two negative values, as indicated in Fig. 3B, and consequently regulates the flow of the two currents through the transistor so that they pulsate at an audio rate. The transistor currents also have pulsations, which occur at the intermediate frequency, as shown in Fig. 3C, and are filtered out by capacitors C2 and C4. These IF filtering currents have been shown in dotted blue.
Since the collector current varies, or pulsates, at the audio frequency being demodulated, an audio voltage is developed across collector load resistor R4. This audio voltage is coupled to the next amplifier stage by capacitor C5, which also serves to "block" the fixed negative voltage of battery M1 and thereby keep it from reaching the next stage. Fig. 1 depicts circuit conditions leading up to a modulation trough. Here the collector current is increasing, and this extra current is drawn from the left plate of C5. This action draws an equal amount of electron current onto the right plate from the external circuit beyond C5 which recognizes this surge of electrons as a positive voltage.
Under the conditions leading up to the modulation peak of Fig. 2, the collector current is decreasing. Thus, the excess electrons being driven through resistor R4 from battery M1 will flow onto the left plate of C5, driving an equal number off the right plate and through the external circuit. The external circuit beyond capacitor C5 recognizes this surge of electrons through it as a negative voltage.
The combination of emitter resistor R3 and filter capacitor C3 is deliberately chosen to have a longer time constant than the duration of one cycle of the lowest audio frequency being de modulated. Consequently, even though the collector-emitter cur rent is pulsating as it comes through the transistor, it is prevented from flowing through R3 in pulsations and thereby developing an audio voltage across R3. Instead, the additional collector current which flows during the modulation trough is shunted momentarily onto the top plate of capacitor C3, and an equal quantity of electrons flows harmlessly from the lower plate into ground.
This is one half-cycle of audio filter current. If there were no capacitor, this extra collector current would have to flow immediately through resistor R3, and, in so doing, would cause an additional component of negative voltage at the upper terminal of R3. Since this is one of the two important biasing voltages of the transistor, a more negative voltage at the emitter would restrict or reduce the flow of both currents through the transistor, just when they were trying to increase. This would be de generation.
The decrease in collector current during the modulation peak of Fig. 2 would likewise cause a smaller voltage drop across R3 and, in turn, a less negative voltage at the emitter. As a result, both currents through the transistor would increase, just when they were supposed to decrease. This would be another half-cycle of degeneration. When a large enough filter capacitor is connected across R3, the excess electrons stored on its upper plate during the previous half-cycle will not be drawn off the top plate, and will flow down through R3 to ground, along with the regular flow of collector current. While this action is occurring, the filter current below C3 will be drawn upward from ground, toward the capacitor. This constitutes the second half-cycle of filter action.
AMPLIFIER WITH AGC DIODE
Figs. 4 and 5 show two successive half-cycles of a single radio-frequency cycle of operation for a fairly typical IF amplifier and detector circuit. The diode also provides automatic gain control (AGC), frequently called automatic volume control (AVC). A fairly common intermediate frequency in broadcast receivers is 455 kilocycles per second. (In electronics terminology, "per second" is understood and so is omitted, and "kilocycles" is abbreviated to kc.) An intermediate frequency is normally much lower than the original carrier or radio frequency, and as such is some what easier to handle-meaning that the circuits which amplify it are less susceptible to losses from parasitic oscillations, radio-frequency interference (RFI) , and other such disturbances.
At the same time, a frequency as high as 455 kilocycles per second is high enough to be classed as a "radio frequency," so that tuned circuits of reasonable size can be put together which will resonate at the basic frequency, with many attendant advantages of signal strength gain, frequency selectivity, etc.
The components which make up this completed circuit are as follows:
R1-Voltage-divider and biasing resistor.
R2-Voltage-divider and AGC resistor.
R3-Voltage-divider and audio-output resistor.
R4-Emitter stabilizing resistor.
C1-Input tank-circuit capacitor.
C2-Output tank-circuit capacitor.
C3-RF bypass capacitor.
C5-Emitter bypass capacitor.
C6-IF neutralizing capacitor.
T1-Input IF transformer.
T2-Output IF transformer.
M2-Battery or other DC power supply.
Identification of Currents There are a great many separate electron currents flowing in this circuit. Each current should be clearly identified in your mind before you can hope to understand the movements, and more important, the functions performed by these currents. First, there are the usual three currents which will flow in the average transistor circuit under "static" conditions, irrespective of whether a signal voltage or current is actually being amplified.
These static currents are:
1. Voltage-divider current ( dotted green).
2. Base-emitter current (solid green).
3. Collector-emitter current (solid red). In addition to these static currents, several additional cur rents come into existence when a signal voltage or current is being amplified. These currents include:
4. Input signal current (solid blue).
5. IF tank current in both the input and output tank circuits (also in solid blue).
6. Output secondary current ( also in solid blue) .
7. IF neutralizing current ( dotted blue) .
8. Unidirectional diode current ( dotted red).
9. Emitter filter current (also in dotted red).
10. AGC current (solid green in Figs. 6 through 9).
Details of Operation
The movements of most of these currents can be understood from Figs. 4 and 5. In order that the actual generation of an audio frequency current and voltage can be visualized how ever, it is helpful to resort to additional circuit diagrams, of a type which will depict audio rather than intermediate-frequency haH-cycles. Also, the generation of an automatic gain-control current and voltage can best be visualized by using extra dia grams to show how circuit conditions change as the signal strength does (signal fade or buildup). Figs. 6 through 9 depict these current and voltage changes. These extra diagrams are necessary because of the three widely separated frequencies that are always involved in the demodulation of an RF or IF carrier signal, and then in the development of a DC voltage which will be proportional to the strength of the original carrier signal. The latter may consequently be used to provide automatic control of the transistor gain and thus of the volume of the audio signal being delivered by the speaker.
These three frequencies are:
1. The carrier or intermediate frequency-in this case, 455,000 cycles per second.
2. The modulation frequency which is carried by the IF carrier and which, after demodulation, becomes the audio frequency from the speaker. A good average audio frequency in the "listening" range of frequencies is represented by the key of middle C, whose pitch is 256 cycles per second.
3. The frequency at which signal fades or buildups will occur because of anomalous propagation conditions. Fades and buildups occur independently of each other, so it is inaccurate to imply that such a thing as a single whole cycle of signal fade and signal buildup exists. It is more accurate to consider individual half-cycles, such as either a fade or a buildup, and to understand within what length of time such an event must occur. A signal fade or buildup may require several seconds or minutes to complete itself.
The time of one IF cycle of operation is always equal to the reciprocal of the frequency, in this case 1/455,000 second.
Thus, one half of one cycle will require slightly longer than one millionth of a second to complete itself. This is about the length of time required for the actions occurring in the tank circuits of Fig. 4. In the tank circuit consisting of the secondary winding of T1 and capacitor C1, the electrons which make up the tank current or circulating current have moved upward through T1 and are amassed on the upper plate of capacitor C1. Thus the voltage across the entire tank circuit has its maximum negative value at the end of this first half-cycle. The amount, or value, of this tank voltage exists to a lesser and lesser degree across portions of the secondary winding. Halfway down this coil, the instantaneous negative voltage at that point will be half of what it is at the top, and so on.
The base of the emitter is connected directly to a point that is quite far down on the secondary winding. This is done for impedance-matching purposes. The concept of impedance match ing is a difficult one to visualize qualitatively. Impedance, like resistance, is a ratio between an existing voltage and the current which this voltage will set in motion. The concept of impedance also represents the ratio of a change in an existing voltage across a circuit, to the change in current through this same circuit as a result of this voltage change.
A transistor connected in a common-emitter configuration such as this one, is said to have a very low input impedance.
This means that the ratio between a change in input voltage and the resulting change in current flow is low. In other words, only a small change in voltage will cause a substantial change in current. As always, in discussing and using the term "impedance," it is not merely helpful but actually mandatory that we under stand exactly which voltage and which current we are talking about.
The input impedance of a transistor refers to the amount of change in voltage necessary between emitter and base in order to produce the desired change in base-emitter current through the transistor. The voltage difference between base and emitter is one of the fundamental biasing conditions of a transistor, and you have seen that this voltage difference is normally only a small fraction of a volt. For the desired fluctuations in base emitter current to occur, it is only necessary to vary this existing base-emitter voltage difference by the tiniest fraction of a volt.
Thus, a circuit in which a very small voltage change causes a substantial current change is a "low-impedance" circuit.
It will be helpful to remember the Ohm's-law relationship between resistance, voltage, and current whenever the impedance of a circuit is under discussion. Impedance, like resistance, is nothing more than a measure of opposition to the flow of electron current. For this reason, it can always be expressed mathematically as a ratio between voltage and current, just like resistance.
Transistors which are used in common-emitter configurations like this one are considered to have a high "output impedance." The output impedance of a transistor is again a means of expressing a ratio between a particular voltage and current. In an output circuit, we are interested in knowing what effect a change in collector voltage will have on the amount of collector-emitter current. Normally, it has relatively little effect. In this respect, the transistor is quite comparable to the pentode vacuum tube, where the plate voltage has only a small effect on the amount of plate current. As a result, the plate circuit of the pentode is considered to have a high output impedance.
For the transistor we can conclude, if we keep the Ohm's-law relationship in mind, that the output impedance is high by recognizing that an average change in the voltage at the collector will produce only a very small change in the amount of collector cur rent. The important biasing conditions of a transistor are the voltages at the base and emitter. The difference between these two voltages exercises an overriding influence on the amount of the two currents which flow through a transistor.
Returning to the input circuit, it is necessary to use only a small portion of the voltage developed across the tank circuit consisting of capacitor C1 and the secondary winding of T1. This is why the inductor is tapped across only a small portion of its length. One might well ask why a tuned circuit is used, when the tank voltage it develops is much higher than is necessary or than can be used for amplification. The answer is that a tuned circuit provides the important feature of selectivity, or discrimination between signals of different frequencies. Depending on its values of inductance and capacitance, a highly tuned circuit will oscillate strongly at one particular frequency and will "reject" all others including those close to its own frequency of oscillation.
The selectivity of a tuned circuit varies in accordance with the Q of the coil which is part of that circuit. Q in this usage refers to "quality"-unlike in Coulomb's law, where Q refers to the "quantity" of electric charge.
The Q of a coil is the ratio between its reactance and resistance and is written arithmetically as:
Q is the coil "quality,"
Q=2 pi fL
f is the frequency of operation in cycles per second,
L is the coil inductance in henrys,
R is the coil resistance in ohms.
This relationship tells us that by increasing the inductance, L, in a coil without changing its resistance, R, we can greatly increase its Q and thus improve the selectivity of the tuned circuit of which the coil is a part.
The "output impedance" of the transistor collector is also "matched" to an appropriate point on the primary winding of output transformer T2. The pulsations of collector current flowing upward through the lower portion of this primary winding will sustain an oscillation of tank-current electrons throughout the entire tank circuit. The exact phase relationships between these two currents have been discussed in greater detail in a preceding Section on the Hartley oscillator. The cases are similar, because in a Hartley oscillator, current is drawn through a small portion of an inductor and, by autotransformer action, in turn supports an oscillation in the entire tank circuit.
The appropriate phase relations between the collector current, tank current, and tank voltage have been depicted in Figs. 4 and 5. In Fig. 4, while the collector-emitter current (in solid red) is increasing in the upward direction through the lower part of the primary winding of T2, it induces another current to flow at an ever-increasing rate in the downward direction through the entire primary winding. This induced current supports the tank current (in solid blue) and is essentially in phase with it. The tank current flows downward through the primary winding of T2 during the negative half-cycle of Fig. 4, and this accounts for the negative voltage shown on the lower plate of capacitor C2 at the end of this negative half-cycle.
During the positive half-cycle of Fig. 5, the collector cur rent is still flowing upward through the lower portion of the primary winding of T2, but is now decreasing. Autotransformer action will now cause an induced current to flow in the same upward direction through the entire winding, but at an increasing rate. This induced current supports the main tank-circuit oscillation, which can be seen flowing upward through the winding in Fig. 5 and delivering electrons to the upper plate of capacitor C2. So, at the end of this positive half-cycle the upper plate of capacitor C2 exhibits a negative charge or voltage.
Probably the most important feature of a tuned tank circuit is that a relatively small amount of replenishment or support can set a sizable amount of electron current in oscillation and maintain it in oscillation. Thus, the current flow induced by the pulsations of collector current is intrinsically quite small in comparison with the amount of tank current which it maintains in oscillation.
This large tank current induces a current (solid blue) at the same frequency in the secondary winding of T2. Whenever alternating current flows through an inductor, an alternating voltage (known by such names as "induced emf," "counter emf," or "back emf") must exist across the inductor terminals. Its instantaneous polarity is always related directly to the direction in which its associated current flows, and also depends on whether this current is increasing or decreasing. (For a fuller discussion of the phase relationships existing between applied and induced voltages and current, refer to the introductory Section of the guide on oscillators in this "Basic Electronics" series.)
The Principle of Neutralization
Neutralization is a technique used for coupling some of the energy from an output circuit of a tube or transistor back to its input circuit for the express purpose of preventing the circuit from breaking into self-sustained oscillations. It is a form of negative feedback between output and input, and is provided to counteract the effects of positive feedback which may be inherent in the tube or transistor. Let us examine the nature of this positive feedback which is inherent within transistors, and then see how capacitor C6 provides the desired negative feedback to neutralize the positive feedback.
The normal flow path for collector current in the PNP transistor is into the collector and out the emitter. In Fig. 4, you can see the increase in collector current following this path.
However, there is inevitably some capacitance between the interior of a transistor and the external wires which lead up to it. Because of these inherent capacitances, the increase in collector current flowing through the base and toward the emitter in Fig. 4 will drive some electron current away from the base and into the external circuit leading up to the base.
This current (in dotted red) flows in a direction that makes the transistor base more negative. This follows from the universal fact that when electron current is moving through a conductor, the terminal or point from which the electrons move is more negative than the point towards which they move.
In Fig. 5 the decrease in the collector current passing through the base again exerts a capacitive effect on the electrons within the wire leading up to the base, this time drawing them toward the base. Since these electrons are being drawn through the entire external circuit between base and ground, they develop a component of positive voltage at the base.
Both of these current actions in the circuit external to the base are classified as positive, or regenerative, feedback be cause they tend to reinforce the very conditions which caused them. During the negative half-cycle of Fig. 4, this component of negative voltage at the base further increases the two currents through the transistor. The resulting additional increase in collector current will make the component of negative voltage at the base still more negative, and so on. During the positive half-cycle of Fig. 5, the small component of positive voltage at the base decreases the two currents through the transistors.
Because of the capacitive effect, the resulting additional decrease in collector current will make the component of positive voltage at the base still more positive.
These positive-feedback actions can be nullified or neutralized by causing another current to flow side by side in the external base circuit with this positive feedback current, but in the opposite direction. This is the neutralizing current, shown in dotted blue, and it can be obtained from the appropriate side of the secondary winding of T2. ( A connection to the wrong end of the coil would give more positive feedback and probably lead to self-sustained oscillations.) In Fig. 4, the top of the secondary winding of T2 is assumed to be at a positive voltage as indicated by the blue + sign. This voltage is "coupled," via neutralizing capacitor C6, to the base of the transistor. Thus, a component of feedback current is drawn toward the transistor base at the same time the undesired feedback current is being driven away from the base.
In Fig. 5, when the top of the secondary winding of T2 is negative, electrons are driven to the left, through capacitor C6, and thus flow away from the base at the same time the un wanted feedback current is flowing toward it. In this fashion, neutralization automatically and continually compensates, during each cycle, for the positive feedback caused by inherent capacitances within the transistor.
The Demodulation Process
Diode M1 in this circuit is a unidirectional device ( one that permits current to flow in only one direction through it). Electrons can flow with relative ease against the direction of the arrow (to the left, in other words), but only with very great difficulty can they be made to flow in the opposite direction.
This inherent property makes the diode a useful rectifying device for the demodulation of a modulated signal. When the top of the secondary winding of T2 has a positive voltage induced on it (such as during the half-cycle of Fig. 4), electrons (in dotted red) will flow up from ground, through resistor R3 and the diode.
This current flow creates a positive voltage at the top of R3, as indicated by the red plus signs on capacitor C4.
During alternate hall-cycles as in Fig. 5, the top of the secondary winding of T2 has a negative voltage induced on it, so no current flows through diode M1 in either direction. The positive voltage on the upper plate of capacitor C4 persists throughout this positive half-cycle, because of the "integrating" action of the long time-constant RC combination consisting of R3 and C4.
Any RC combination is a long time-constant combination to a particular frequency if the resistance in ohms multiplied by the capacitance in farads is more than five or six times greater than the time required for a single whole cycle of current movement to occur at the particular frequency. This is in accordance with the time-constant formula:
T is the time constant of the combination in seconds,
R is the resistance in ohms,
C is the capacitance in farads.
If an unmodulated carrier signal were being received, every RF cycle would have the same strength, and the voltage on the upper plate of C4 would be essentially DC. When a modulated signal is being received the strength of the individual RF cycles is not constant. Rather, it varies in accordance with modulation, and so does the voltage on C4. When a strong RF cycle is being received (this happens during a modulation peak), the induced voltage across the secondary winding of T2 will be higher. Thus, more electron current will be drawn upward through R3 and the diode. This means that a higher positive voltage will exist across R3 and on the upper plate of C4.
When a weak RF cycle is being received (this happens during a modulation trough), the induced voltage across the secondary winding of T2 will be lower. Now, less electron current will be drawn upward through R3, and a lower positive voltage will exist across R3 and on the upper plate of C4. Since the strength of the individual cycles of RF depends on the amount of modulation, the voltage produced at the top of resistor R3 is an audio frequency voltage.
Figs. 6 and 7 show two successive half-cycles of the audio voltage as it appears for the first time, immediately after demodulation, in a typical receiver. Circuit components in these figures have been numbered to correspond to their counterparts in Figs. 4 and 5. The radio-frequency currents have not been shown in Figs. 6 and 7. These illustrations depict one entire audio cycle, consisting of a modulation trough followed by a modulation peak, while a weakened carrier signal is being received. The audio currents are shown in dotted red, the same as in Figs. 4 and 5. The AGC current (so labeled, and shown in solid green) flows in either direction along part of the same path used by the voltage-divider current (in dotted green in Figs. 4 and 5) . During a modulation trough, when a low positive voltage exists on the upper plate of C4, the AGC current is drawn to the left, through resistor R2, by the higher positive voltage stored on the upper plate of C7.
During a modulation peak as in Fig. 7, this AGC current is drawn to the right, through resistor R2, by the higher positive voltage now stored on the upper plate of capacitor C4.
The intrinsic amount of positive voltage stored on the upper plate of AGC capacitor C7 does not change during a single audio cycle, even though electrons flow onto it during a modulation trough (Fig. 6) and out of it during a modulation peak (Fig. 7) . The reason is that the combination of resistor R2 and capacitor C7 forms a long time-constant to the lowest audio frequency likely to be encountered. This means these components are large enough that their product is several times the period of a single low-frequency audio cycle. When these components are made large enough, the amount of electrons which flow through R2 on successive half-cycles will be insignificant, com pared with the number of positive ions already stored on the upper plate of C7. Thus, by appropriate choice of the sizes of these two components, a DC voltage can be developed on the upper plate of C7 that does not vary with the modulation represented by the audio signal.
The voltage stored on an AGC capacitor such as C7 will always adjust itself to the average value of the trough and peak voltages on the upper plate of the audio-output capacitor in this case, C4. This average value will remain the same unless the over-all strength of the RF carrier changes because of propagation anomalies, which cause what are known as signal fades or buildups. Figs. 8 and 9 depict two successive audio half cycles during a period of excessive carrier-signal strength, or what is called a signal buildup. All the radio-frequency tank currents shown in solid blue in Figs. 4 and 5 will become proportionately stronger during such a period. As a result, more audio current will be drawn upward through R3 and diode M1 during both a modulation peak and a modulation trough. This is depicted by the additional dotted red lines in Figs. 8 and 9.
This additional current flow increases the trough and peak voltages stored on C4 and so, increases their average value.
The voltage stored on capacitor C7 is "replenished" by the voltage stored on C4. During a signal buildup, when both the audio modulation trough voltage (Fig. 8) and the audio modulation peak voltage (Fig. 9) are increased, their average value is also increased. Consequently, the positive voltage on the upper plate of C7 must increase to this average value. This important action is accomplished by the AGC current shown in Figs. 6 through 9. During a period of normal or unchanging signal strength such as are shown in the two half-cycles of Figs. 4 and 5, an AGC current will flow, at the audio frequency, back and forth between capacitors C7 and C4, through resistor R2. The amount of current flowing to the left during each modulation trough will be exactly equal to the amount flowing to the right during the next modulation peak. Because of this, the positive voltage stored on C7 is maintained at the average value of the high and low positive voltages on C4.
When signal strength is increased by a signal buildup, both the high and low positive voltages will be increased in value.
When this happens, the two half-cycles of AGC current flowing through resistor R2 will become "unbalanced." In other words, more electrons will flow away from C7 during the modulation peak of Fig. 9, and fewer electrons will flow into C7 during the modulation trough of Fig. 8. This condition of current unbalance will continue until the AGC voltage stored on C7 becomes sufficiently positive to just equal the average value of the trough and peak voltages on C4.
The circuit actions involved in achieving automatic gain control in a transistor circuit are fundamentally similar to those in vacuum-tube circuitry. These principles have been discussed with greater detail in the Section on automatic volume control in the guide about detector and rectifier circuits in this "Basic Electronics" series. Also, waveform diagrams appearing in that book may help to clarify the meaning and significance of terms such as modulation trough, modulation peak, signal fade, and signal buildup. The most important difference between A VC circuitry using tubes and transistors is that in tube circuitry a negative A VC voltage normally is developed whereas in the example of this Section, the A VC voltage developed on capacitor C7 is positive instead.
How AGC Voltage Controls Transistor Gain
In this type of circuit, the AGC voltage controls, or regulates, the transistor gain by exercising some control over the base emitter current which flows through the transistor. Since this is a PNP transistor, any positive component of voltage applied to the base of the transistor will decrease the base-emitter current (called "biasing" current). Its quantity depends on the biasing conditions (meaning the biasing voltages) at the base and emitter. As was pointed out earlier in this Section, a certain negative voltage is applied to the emitter of X1 (as a result of current flow through resistor R4), and a slightly higher negative voltage is applied to the base ( as a result of voltage divider action through R1, R2, and R3). The difference between these two applied voltages is one of the fundamental biasing conditions of the transistor (the most important one, in fact), and it controls the amount of electron current flowing from base to emitter (the base-emitter biasing current). The existence of a permanent positive voltage on the upper plate of capacitor C7 will reduce the negative voltage created at the base of the transistor by the flow of voltage-divider cur rent through R1, R2, and R3. The positive AGC voltage shown on C7 is not a composite value of all the voltages at the base of the transistor, but only the result of the AGC filter action occurring between R2 and C7. You have already seen that four other voltages exist at the base, each contributing in some small degree to the amount of base-emitter biasing current flowing.
The AGC voltage makes a fifth one. These five voltages are:
1. The voltage resulting from voltage-divider current through resistors R1, R2, and R3. This is a fixed, or DC, voltage which is negative at the junction of R1 and R2, and consequently negative at the base.
2. The signal voltage, which is alternately positive and negative at the intermediate frequency, and which therefore lowers and raises the negative voltage at the base.
3. The voltage resulting from positive feedback caused within the transistor by capacitance between the internal elements (B, C, and E) and the external wiring of the transistor (IF) .
4. The voltage due to negative feedback from output circuit to input circuit, through capacitor C6. (IF)
5. The AGC voltage on C7. Although essentially a DC voltage, it does vary if and when the signal strength of RF carrier varies.
The combination of resistor R2 and capacitor C7 is chosen so that their product will be a "long time-constant" when com pared with the duration of a single low-frequency audio cycle.
If the lowest audio frequency being amplified were 50 cycles per second, with a period of .02 second for each cycle, the R2-C7 combination would be a "long time-constant" combination to this frequency if their product were more than five times longer than this period--i.e., a tenth of a second or longer.
The Audio-Output Voltage
An audio-output voltage is developed across resistor R3, by the pulsating DC which continuously flows upward through it.
This current (in dotted red) also flows through M1, and as a result of the demodulation process previously described, its pulsations occur at the audio frequency. The arrow pointing into R3 indicates that this is a potentiometer, or variable resistor, with which any desired fraction of the total voltage may be tapped off for amplification by the next succeeding audio-amplifier stage. Potentiometer R3 thus constitutes a manual volume control for the receiver in which it is installed.