A brief summary of the differences between the two systems of broadcast--AM and FM--will aid in the understanding of FM receiver circuits. As we have done in other sections, only those points of theory necessary to the understanding of the troubleshooting techniques will be discussed.
In an AM transmitter, the audio modulation is mixed with the RF carrier in the modulated amplifier stage. As shown in Fig. 12-1, this produces three signals which are radiated from the antenna:
1. The original carrier.
2. The sum of the carrier and the audio, called the upper sideband.
3. The difference between the carrier and the audio, called the lower sideband.
Fig. 12-1. AM modulation.
The spacing between the main carrier and either sideband represents the frequency of the audio modulation. The amplitude (volume level) of the sound is represented by the relative amplitude of the sideband compared to the carrier.
The term 100 % modulation is used when the total amplitude of both sidebands added together is equal to one-half of the transmitter power. These relationships are shown in Fig. 12-2, which represents the transmitted signal from a station operating at 620 khz and modulated with a 2000- hz frequency. Note that the total bandwidth is 4000 cycles. The FCC has limited the bandwidth of AM broadcasting stations to a maximum of 10,000 cycles. This means that the maximum audio signal allowed is 5000 cycles.
Thus, it can be seen that AM stations are limited in two ways:
1. The maximum audio power in the sidebands must not exceed one-half of the total transmitter power.
2. The maximum audio frequency must not exceed 5000 cycles per second.
Fig. 12-2. Carrier and sidebands at 100% modulation.
Fig. 12-3. FM modulation.
Only one of these limitations applies to an FM station, how ever; the maximum audio power is limited, but the maximum frequency of the audio is practically unlimited. In fact, an FM station can be modulated by extremely high frequencies, even in the range of supersonic frequencies.
In an FM transmitter (Fig. 12-3), the audio is applied directly to the carrier-generating oscillator, where it is used to change the frequency of the carrier. (Technically, there is mixing of the audio and RF to produce sidebands, but these are not the intentional result of the modulation, and they are not used to carry the audio signal.) It is especially important to understand that an increase in the amplitude (volume level) of the modulation at the transmitter results in a greater deviation of the carrier from its center frequency, but the carrier amplitude does not change.
A change in the frequency of the modulating audio does not affect the amount of deviation but only changes the speed at which the deviation takes place.
Fig. 12-4. Modulation of an FM carrier.
Fig. 12-4 illustrates the changes in the frequency of the FM carrier for various kinds of audio. The movement from side to side represents changes in frequency of a transmitter carrier corresponding to changes in amplitude of the audio modulation. The carrier amplitude is always 100 % , which means there are no appreciable changes in the FM carrier amplitude with changes in the modulation. The carrier does change in two other ways, representing changes in frequency and amplitude of the audio, but in neither case does the carrier amplitude change.
By comparing the two methods, we find that FM has several advantages:
1. The bandwidth is not directly affected by the frequency of the audio.
2. The transmitter is modulated in a low-power stage and thus requires very little modulating power to control a high-power RF carrier.
3. The amplitude of the carrier does not change.
4. The limitation on the FM station by the FCC is in terms of the maximum bandwidth allowed, and this primarily affects the volume level and not the frequency of the modulation. Thus, the FM station can transmit any audio or supersonic modulation as long as the volume-level deviation is kept within the limits specified by the FCC.
5. The receiver can be made immune to noise because most noise is caused by changes in amplitude, and the receiver does not need to respond to amplitude changes.
The FCC places a limitation of 150 khz as the maximum deviation allowed. This is a deviation of 75 khz on either side of the center frequency. This allows more than enough dynamic range in volume level. By comparison, television sound (which is FM) is allowed to deviate only ± 25 khz.
In recent years the FCC has allowed some FM stations to increase their deviation to higher bandwidths. This is reason able, since at the high RF frequencies where the FM stations are located, 200 khz is only about 0.2 % of the entire FM band.
In AM, 200 khz is about 20 % of the entire broadcast band.
Fig. 12-5. Block diagram of an FM receiver.
Fig. 12-6
FM RECEIVER CIRCUITRY
The block diagram in Fig. 12-5 indicates some new stages that are not used in AM receivers-the limiter, the FM detector, the AFC, and the RF amplifier.
A typical limiter is shown in Fig. 12-6. This stage is the last IF amplifier, V7, which is a type 6A U6. The purpose of this stage is to remove all amplitude variations in the incoming signal. In this way, most of the noise and interference will be eliminated without affecting the frequency changes. The input and output waveforms indicate that amplitude variations are removed by the clipping action of the stage, which does not affect the frequency changes. Since the audio sound from the station is represented by the amount of frequency change and the rate at which the frequency changes take place, limiting the amplitude of the signal does not affect the FM sound. All forms of noise are amplitude changes, and these are removed by the limiter.
The limiting action is accomplished by operating the stage with very low plate and screen voltages. These low voltages result in a rather low amplitude of input voltage, causing the output signal to reach the maximum possible amplitude. This condition is called saturating the limiter, and once this state is reached, no further changes in the amplitude of the output are possible. The input signal must, of course, be great enough to drive the limiter stage to saturation before limiting will take place. This means that there is no limiting action on very weak carriers. It is for this reason that noise is sometimes heard on supposedly "noise-free" FM receivers.
Because of the large grid resistor R25 and grid capacitor C42, a small amount of grid-leak bias is developed in the limiter stage. Besides aiding the limiting action by reducing the sensitivity of the stage, this voltage varies in proportion to the strength of the incoming signal and serves as an AVC. In FM receivers, A VC is usually applied only to the RF amplifier in the tuner. We shall see later in this section that there are other sources of AVC voltage in the FM receiver which may be used.
The Tuner The tuner is usually a separate unit housed on a subchassis which can be removed or replaced by removing a few screws and unsoldering a few connections. The dotted lines appearing in the schematic indicate the subchassis assembly.
The circuits shown in Fig. 12-7 are very simple since no A VC or AFC is used. Only a single twin triode is employed, and a minimum of tuned circuits are included. In part A of Fig. 12-7, the input of the RF amplifier is fixed-tuned. The plate circuit is tuned, along with the oscillator tank, by the station selector.
The signal from the RF amplifier is coupled to the grid of the mixer through CS. C7 and C9 aid in keeping the tracking aligned as the receiver is tuned through the FM band (88 mhz to 108 mhz). The reactance of all three capacitors changes with frequency, but since the capacitors are connected in the circuit differently, the changes work together to keep the tracking nearly perfect.
The oscillator uses feedback from the plate, through C10 and the primary winding of the oscillator transformer. The secondary of the transformer is in the grid circuit of the oscillator, and this produces the proper feedback to sustain oscillation. The oscillator signal is mixed with the incoming RF station carrier at the grid of the mixer/ osc stage.
Fig. 12-7. Typical FM tuners.
Since the oscillator is always tuned 10.7 mhz higher than the incoming station, a beat is developed in the mixer at this IF frequency. The plate transformer of the mixer is tuned to 10.7 mhz.
R3, R4, and C14 form the B+ decoupling network. These resistors are of particular importance to the troubleshooting technician because they usually burn out when the tube shorts.
A defective tuner is easily isolated when it is found that an IF (10.7-mhz) signal can be injected at the first IF stage but not at the grid of the mixer. Replacing one or both resistors is a simple repair job.
FM circuits usually include decoupling of the filament lines going to the tuner and occasionally to the IF stages. The powdered-iron sleeve in conjunction with C6, in the circuit in part A of Fig. 12-7, is for this purpose. Although the components in the decoupling networks seldom fail, the technician should be aware of the possibility of damage to the power supply due to a shorted unit such as C6.
The circuit in part B of Fig. 12-7 uses a different tube, the ECC85, which is equivalent to the more common 6AQ8. The plates of the tubes in this circuit are shunt fed. Notice, for example, that coil L3 feeds B+ to pin 1 of V1, but that the resonant tank circuit is the one containing L4, and is isolated from B+ by C7. Therefore, no plate current flows through the tuned circuit, allowing one side of the coil and capacitor to be grounded.
The output of the RF amplifier is tuned to one particular FM station, and is coupled to the grid of the mixer from the center tap of the RF plate tank. This signal is fed through C9 to the secondary of the oscillator tank where it is mixed with the oscillator signal, the combination appearing at pin 7 of the mixer. The plate circuit of the mixer is tuned to the IF frequency and is also shunt fed so that plate current does not flow through L9.
The circuit includes C18 to bypass the filament line, and R9 and C10 which form the B+ decoupling network. The schematic symbol for C10 is often found in VHF circuits and indicates that the capacitor is in the form of a feed-through insulator through which the wire passes from the underside of the chassis to the top side.
Some of the other capacitors in the circuit are specially constructed to give them a specific temperature coefficient, which means that the change in capacity with a change in tempera ture is controlled and can be predicted. These capacitors must always be replaced with similar units so that the balance of changes throughout the circuit as it heats will be maintained.
Failure to compensate for changes in temperature causes drifting of the tuning with resulting distortion. Grid blocking capacitors, such as C8 and C9 in the circuit in part A of Fig. 12-7, are the most likely components in the receiver to cause drifting.
Automatic Frequency Control (AFC)
AFC has been advertised as a special circuit which makes receiver tuning easier because it eliminates the possibility of setting the tuning dial slightly to either side of the carrier center frequency. AFC does this superbly, and it also compensates for drifting by automatically adjusting the oscillator frequency to keep the carrier centered in the detector at all times.
Until recent years, the method of achieving automatic control of the oscillator frequency was to employ a reactance tube across the oscillator tank. V2A in Fig. 12-8 is connected in this manner, while V2B is the oscillator tube itself. The conduction of the AFC triode creates an effective reactance across the tank that is dependent on the amount of bias applied to the grid through the control line. The control tube acts as a variable reactance (in this case capacitive) across the oscillator tank.
When the oscillator drifts slightly, the incoming carrier will no longer be centered perfectly in the IF bandpass of the receiver, and this reduces the receiver's response to deviation in one direction, causing distortion. The discriminator develops a voltage at pin 5 of V10 which is proportional to the amount by which the carrier has shifted from the center of the bandpass.
Fig. 12-8
The way the reactance tube responds to this voltage to correct the oscillator frequency is illustrated by the simplified drawing in Fig. 12-9. The B+ circuit to the plate of the reactance tube has been eliminated because it plays no part in the reactance function of the tube. Also, C11 is omitted because at these frequencies its effect is negligible. R9 and R10, the fixed cathode-bias divider, is also omitted for simplification.
The resistance between grid and cathode is taken to be 220 ohms, the value of R7. This is approximately correct when only the AC path is considered.
The oscillator voltage appearing across the oscillator tank, L5, is designated E1• C12 and R7 are connected across this voltage and the reactance of C12 is much greater than R7, making the series path mostly capacitive. When AC current is passed through a network which contains more capacitive reactance than inductive reactance, the current will lead the source voltage, and when the network is mainly capacitive, the phase angle approaches 90°. Therefore, 11 (which flows through the series network, C12 and R7) leads E1 by nearly 90°. Voltage Ex, produced in R7 by this current, is in phase with the current. This means that Ex is also leading E1 by 90°.
Fig. 12-9. Simplified AFC circuit.
Since Ex appears between the grid and cathode of the reactance tube, it controls Ip, the AC plate current of the tube.
Now, if Ip follows faithfully the variations of Ex, then Ip will also lead E1 by 90°. * When Ip passes through the oscillator tank, it is exactly the kind of current which would appear in the tank if a capacitor were added in parallel with C8. The amount of Ip depends on the DC grid bias supplied by the control voltage which, in turn, is proportional to the amount of detuning due to oscillator drift. In this manner, the oscillator can be kept precisely on frequency, regardless of changes in voltages or components.
Ordinary crystal diodes can be used for AFC, as shown in Fig. 12-10. The operation is quite simple, since no reactance tube is employed. The grid-leak resistor is divided into R1 and R2 so that a part of the grid-leak voltage can be placed on the diode plate. This negative voltage keeps the resistance of the diode very high, and the effect of C1 across the tank is mini mum. When the control voltage applied at the cathode is more negative than the fixed bias, the resistance of the diode is reduced, and C1 is more effectively across the tank. Thus, the AFC system consists essentially of the shunt capacitor, C1, in series with a variable resistance.
* The 180° phase shift through a common cathode circuit applies to phase relations between grid voltage and plate voltage. Here, we are comparing grid voltage and plate current.
Fig 12-10. AFC circuit using a crystal diode.
Fig. 12-11. AFC circuit using a Varicap.
Another modern form of AFC which has become very popular uses a special diode known as a varicap. This unit presents an amount of capacitance in the circuit depending on the voltage applied across it. Fig. 12-11 shows a modern tuner which uses only one dual-triode tube to perform the functions of RF amplifier, mixer, oscillator, and AFC. The circuit employs a varicap.
THE FM DETECTOR
To recover the audio modulation from the transmitted signal, it is necessary to have a circuit whose output is proportional to the amount by which the carrier deviates from the center frequency. There are three basic circuits for accomplishing this task. They are:
1. The Foster-Seely discriminator.
2. The ratio detector.
3. The gated-beam, or locked-oscillator, detector.
The first two are found most often in FM receivers, and the third is used primarily in TV sound sections. A brief discussion of the operation of each will aid in understanding the alignment procedures to be described.
The Discriminator
Fig. 12-12. A discriminator.
In Fig. 12-12, the discriminator is shown in its most basic form, with the incoming carrier voltage designated as Ee across the primary of the input transformer. The coil, L1, connected between the center tap of the secondary and the junction of R1 and R2, is not found in practical circuits, but is useful here to give an indication that the full carrier volt age, Ee, appears in series with each diode. This is true because of the capacitive coupling between the top of the primary and the center of the secondary. While it is somewhat of an over simplification to say that the Ee which is across the secondary and the primary are all exactly the same, they differ only in phase. The differences seem to balance out, so they do not affect this explanation.
When the carrier is in the center of the IF bandpass (when there is no modulation), a current, l 0 , is induced in the resonant tank circuit formed by the secondary and C1. Taking Ee across the secondary as a reference, the phase relations are as shown. The two voltages, e1 and e2, formed by the center-tap must be 180° out of phase since they are at opposite ends of a coil, and they must also be 90° out of phase with l 0. This 90° phase difference arises from the fact that the current and voltage in any coil are always 90° out of phase unless there is a condition of resonance. Since e1 and e2 are not taken across the entire coil, the condition of resonance which exists between the full secondary and C1 does not apply to these voltages; they are merely the voltage resulting from the current, le, flowing through a random inductance.
Now, giving attention to the cathode end of the circuit, we see that in the upper diode the voltages E0 and e1 appear in series across R1, and in the lower half of the circuit, E0 is in series with e2 across R2. Because of the diodes, current must flow from left to right in the center leg of the circuit, and from right to left in the outside legs. This gives opposite polarities to the voltages across R1 and R2, as shown. The output is the sum of these voltages and will be zero when 11 and hare equal.
The vector diagram shows that the voltages must be added to E0 vectorially because they are not in phase. The fact that the two resultants are equal is also evident, and of course the output voltage taken across R1 and R2 is zero when the carrier is at the resonant frequency of the secondary tank circuit.
The circuit action when the carrier is above resonance is shown in the other vector diagram. l 0 can be seen to have moved away from E0. This is because when the secondary tank is out of resonance, the current and voltage (10 and E0 ) in the tank are not in phase. When the frequency is higher than resonance, the circuit is inductive and the current lags the volt age.
When l 0 moves, voltages e1 and e2 which it generates must also move so that they can stay 90° out of phase with l 0 • With these new phase relations, the vector sums across R1 and R2 are no longer equal. This means that the opposite polarities of these voltages no longer cancel, and an output voltage now appears.
From the vector diagram it can be seen that E0 and e2 are closer together in phase and thus produce a resultant of greater magnitude, indicating that the voltage across R2 is increased. However, e1 is now farther away from Ee, and the sum of these two is reduced, indicating that the voltage across R1 is less than before.
The output voltage reverses in polarity when the carrier swings below resonance and, in this manner, an AC output is developed which has a frequency corresponding to the rate of carrier deviation, which is the frequency of the modulating audio. Further, the amplitude of the output voltage corresponds to how far the carrier moves off center, and this is the amount of deviation corresponding to the amplitude of the audio modulation.
The discriminator circuit has the disadvantage that noise pulses which affect the magnitude of the input voltage, Ee, will also affect the output, since it is partly composed of Ee. The circuit must therefore be used with a limiter stage to remove all amplitude variations. This is inefficient because of the necessity of amplifying all signals up to the level where they saturate the limiter.
The Ratio Detector
A circuit which is more popular than the discriminator is shown in Fig. 12-13. This is a ratio detector and can be distinguished from the discriminator by the fact that the diodes are connected in series; that is, one plate and one cathode are connected to the secondary. Analysis of this circuit will indicate that changes in amplitude of Ee do not appear in the output.
With the diodes in series, the input voltages cause the cur rent, Ii, to flow around the outside loop-through the entire secondary, the diodes, and C1 and C2, all in series. 11 charges C1 and C2 so that the peak carrier voltage appears across the pair of capacitors.
When the carrier is centered exactly on the resonant frequency of the transformer secondary, the voltage applied to D1is the vector sum of Ee and e1. This same voltage is applied to C1. The vector diagram on the left shows that the voltage applied to D2 and to C2 is equal in magnitude, being the vector sum of Ee and e~. The carrier voltage appears across the combination of C1 and C2, and also across the combination of R1 and R2. Because of this, C3 is charged to the peak carrier voltage. It is very important to note that the voltage across C3 is not the output voltage corresponding to audio modulation.
The output signal is developed between the junction of the two capacitors and the junction of the two resistors. When the carrier is centered, the charge on C1 represents one-half of the carrier voltage, and the voltage across R1 is also one half of the carrier voltage. Under these conditions, there is no difference in potential across the output terminals and, consequently, no output. This is to be expected when the carrier is in the center of the bandpass.
Now it can be seen why the ratio detector does not respond to changes in the amplitude of the carrier. Suppose the value of Ee is suddenly increased in amplitude because of a noise pulse but it does not change in frequency. The voltage across C3 cannot change very fast because of the long time required to charge or discharge this large capacitor. Also, the voltage relationships still exist between C1, C2, R1, and R2, whereby there is no output due to equal voltage values at the output terminals. There will be no output voltage, regardless of the value of Ee, unless there is a change in frequency producing unequal voltages on C1 and C2.
Fig. 12-13. A ratio detector.
When the carrier shifts frequency under modulation, the vector sum of Ee and e1 (which charges C1) will no longer equal the vector sum of Ee and e2 (which charges C2). This can be seen in the vector diagram on the right, and the explanation is exactly the same as for the discriminator. The voltages across R1 and R2 remain the same, however, since the charge on C3 is unchanged.
Referring now to the partial schematics in Fig. 12-13, sup pose that the original voltages at resonance were as shown in part A. When the carrier shifts above resonance, the sum of e2 and Ee is larger, and the total of 10 volts might distribute itself across C1 and C2 as shown in part B, with 7 volts across C2 and 3 volts across Cl. There is now a difference in potential across the output terminals because the voltages across the resistors have not changed.
For convenience in the design of the next stage, it is common to take the output relative to one end of C3, as shown in part C. It does not matter which side of C3 is grounded. All that is necessary is to connect one output terminal to a fixed voltage and the other terminal to the voltage at the junction of C1 and C2.
The Gated-Beam Detector
A special tube, the 6BN6, is used in a gated-beam detector.
A variation of this circuit, called the locked-oscillator, quadrature-grid detector, uses a 6DT6. The circuits are nearly the same in principle.
The grids connected to pin 2 and pin 6 are arranged in a special way so that either one is capable of completely cutting off the tube. Also, no current can flow in the tube unless both of these grids are above the cutoff voltage. The tube, therefore, has two gates, either one of which can reduce the plate current to zero.
The plate-current characteristics of the tube are such that a small signal applied to the grid at pin 2 drives the plate cur rent to saturation. This feature enables the stage to perform the function of limiting and thus remove amplitude variations in the signal. After the plate current is driven to saturation, further increases in signal voltage do not change the amplitude of the output signal. As long as the incoming carrier is considerably larger than the voltage required to reach saturation, then a decrease in signal voltage is not likely to be great enough to bring the plate current out of saturation. This means that an amplitude change will not be reproduced in the plate circuit.
Fig. 12-14. A gated-beam detector.
The operation of the circuit as an FM detector is simple and can be seen from a study of Fig. 12-14. The oscillator signal supplied to the quadrature grid (pin 6) is induced in the resonant circuit connected to the grid by capacitive coupling through the tube from the signal on pin 2. Thus, the oscillator sine wave is 90° out of phase with the incoming signal.
The shape of the plate-current pulses is the result of the two special characteristics of the tube, which have been explained. Plate current can flow only when both grids are positive, and these small segments of time correspond to the positions of the sine waves indicated in the diagram. Also, the plate current is driven to saturation very rapidly, and this accounts for the rectangular shape of the pulses.
The action when the incoming carrier shifts frequency under modulation can be seen in part B of Fig. 12-14. The incoming signal on pin 2 is seen to be shifted in phase so that the relationship between the two sine waves is not the same.
This results in plate pulses of different length. It is in this manner that the output signal changes with changes in the frequency of the input signal.
It is natural to ask at this time why the quadrature signal on pin 6 does not also change, since it is generated by the incoming signals. It actually does change, but not to the same extent, so that the two signals do not remain exactly 90° out of phase.
A resistor and capacitor, R36 and C23, in the plate circuit are used to integrate the rectangular pulses and provide an audio signal dependent on the average current contained in the pulses as they vary in width.
The gated beam is not very popular in FM receivers, al though it is used extensively in TV. The reason is that its output is likely to contain a large square-wave component which causes a buzz. Varying the cathode resistor to obtain the best possible limiting is the only way this can be eliminated.
AFC Control Voltage
A DC voltage which varies with the carrier amplitude is present in the circuits of Figs. 12-12 and 12-13. If the oscillator in the tuner drifts so that the incoming station is not centered in the IF bandpass of the receiver, there will be a change in carrier amplitude at the FM detector, and this DC voltage will change accordingly. We will see more clearly later why this causes distortion.
In Fig. 12-12, this voltage is between the junction of R1, R2, and ground. In this discriminator circuit, the output signal changes polarity and amplitude according to the difference in the voltages across R1 and R2. However, the voltage between R2 and ground is always present and always has the same polarity, but its amplitude varies with the strength of the in coming signal.
A similar voltage in the ratio detector of Fig. 12-13 is found across C3. Although this voltage does not respond to sudden changes in carrier amplitude, it does change gradually if the carrier amplitude changes. These two voltages can be taken in either polarity with respect to ground and, when filtered, they serve to control the oscillator through a reactance tube or diode, as explained earlier.
FM RECEIVER ALIGNMENT
Alignment of FM receivers is critical because it is necessary to achieve a relatively flat response throughout a range of 200 khz or more in the IF channel. It is also essential that the response falls off evenly on both sides of center to avoid attenuating the carrier when it deviates in either direction.
Receiver circuits vary so greatly that only a few very general principles can be given. Alignment instructions must be followed exactly as to frequencies, methods of applying the signal generator, and measuring the output. The purpose of the discussion that follows is to help understand these instructions.
Symptoms of Misalignment
Loss of gain and sensitivity indicates a need for IF and tuner alignment when tubes, voltages, and antenna conditions have been checked. Another definite indication of the need for alignment is distortion which seems to change with tuning.
If it is difficult to tune a station so that it falls in the center of the bandpass with room to deviate in both directions, this may mean that the IF channel is too narrow. More likely, however, it means that the secondary of the detector trans former is not tuned to resonance at the IF frequency.
Special Instructions for Alignment of FM Discriminators
Set the generator for a 10.7-mhz unmodulated output, and connect it to the grid of the last IF stage through an isolation capacitor. Connect the VTVM to the center of the two resistors between the cathodes of the diodes. The meter should be set to read negative volts on the 10-volt scale. Turn the equipment on and adjust the attenuator of the signal generator so as to produce a minimum signal which will give an adequate reading on the meter.
Adjust the primary tuning of the transformer, shown in Fig. 12-15 as CD, for a maximum peak reading on the meter.
The primary is usually the top slug, but reference should be made to the manufacturer's data for such information, when ever possible. Move the meter probe to position 2, which is the discriminator output. Zero-center the meter, and again set the generator for minimum signal input consistent with good meter deflection. Adjust the secondary of the transformer, shown in the figure as [2], for a zero reading on the meter. This adjustment is critical and must be made as care fully as possible. A lower-scale reading on the meter can be used for the final touching up. Whenever the secondary is out of resonance, deflection on the meter will be to the positive or negative side of zero-center. Perfect resonance is indicated by zero output from the discriminator with the input carrier signal at exactly 10.7 mhz.
Fig. 12-15. Discriminator alignment.
Fig. 12-16. Ratio-detector alignment.
An interesting check on the accuracy of alignment of the transformer secondary and overall efficiency of the detector is to turn on the AM modulation from the signal generator. If the secondary is perfectly tuned, the audio tone will almost completely disappear, indicating that the detector is least sensitive to AM modulation at this time.
Special Instructions for Alignment of the Ratio Detector
The generator is connected in the circuit in the same way that it was for the discriminator, but the VTVM has a different hookup. Fig. 12-16 shows, in dotted lines, two 100K resistors connected across the capacitor. Since the electrolytic capacitor is grounded, it is necessary to add these resistors if they are not already in the circuit, in order to make the zero-center adjustment. The circuit shown in Fig. 12-16 is the most common type of ratio detector, and it will usually be necessary to add the resistors while the zero-centering is being done. Note that when the resistors are used the negative meter lead is on the audio output.
Fig. 12-17. Alignment of a locked-oscillator, quadrature-grid detector.
With the VTVM connected to position 1, tune the primary of the transformer for a maximum peak reading on the meter.
Then move the meter probe to position 2, adding the resistors to the circuit if necessary. The meter should be zero-centered as before, and the secondary tuned in the same manner as it was for the discriminator.
Special Instructions for Alignment of the Gated-Beam and Locked-Oscillator, Quadrature-Grid Detectors
These two circuits, shown in Figs. 12-17 and 12-18 respectively, are very similar in operation. One exception, however, is that a conventional sharp-cutoff pentode is used in the locked-oscillator circuit to reject amplitude variations, and so does not require a buzz or quieting control. Successful alignment of either circuit depends on knowing that the input transformer must be tuned for maximum output on AM modulation. This can only be done if the input signal is weak enough so that limiting action does not take place.
Short out the quadrature coil, and connect a signal genera tor, set to 10.7 mhz and with a low percentage of AM modulation, to the IF grid. Decrease the output from the generator until there is a noticeable drop in the audio output from the speaker, or until the needle drops on the VTVM connected at the output of the audio power amplifier. This drop in audio indicates limiting is no longer taking place. Adjust the primary and secondary of the input transformer for maximum output, and continue to lower the input signal so that limiting does not take place.
Fig. 12-18. Alignment of a gated-beam detector.
Fig. 12-19. IF stage frequency response.
After the input transformer is peaked, remove the short from the quadrature coil and tune in a strong station on the receiver. Then tune the coil for best audio output and mini mum buzz. If there is a buzz or quieting control in the cathode to reduce the 60-cycle buzz, it should be adjusted at this time.
A slight readjustment of the quadrature coil may be necessary after this operation. Complete alignment of all IF stages and the tuner may be further required in order to obtain undistorted sound of good volume.
Special Instructions for Alignment of IF Stages
As mentioned before, it is not likely that IF amplifiers will require alignment on a regular service job, and the task should never be attempted unless the technician is experienced and has the equipment and manufacturer's data available.
An interesting check of the width of the IF channel can be made with a regular signal generator and a VTVM. First, complete the alignment of the detector stage, and then prepare to take a number of meter readings which will be plotted on a graph. Couple the generator loosely to the mixer stage or to the grid of the first IF stage. The zero-centered VTVM is connected to the output terminals (position 2, in the previous alignment instructions) and the signal generator set at 10.7 mhz. Record the voltage at this point, which should be zero.
Now, adjust the generator in 10-khz steps downward for 200 khz, recording the voltage at each step. Return the generator to 10.7 mhz and repeat the 10- khz steps, this time upward for 200 khz, recording the voltage at each step.
When the voltages are plotted, the resulting curve indicates the total bandwidth of the IF and detector circuits. Fig. 12-19 shows an ideal curve. It is not likely that many receivers can actually produce a curve so wide and linear as this one. But what is important in reducing distortion is that the voltages be equal at points equidistant from the center. For example, if the voltage at 20 khz below the center (at 10.68 mhz) does not agree with the voltage at 20 khz above the center (at 10.72 mhz), this means that there is more audio produced on one side of the deviation than on the other, and this is not faithful reproduction of the transmitted audio.
Another observation can be made from the curve. When the curve turns toward zero (indicating decreasing output) at frequencies too close to the center, this indicates that the IF channel is too narrow, and both fidelity and dynamic range will be affected. 100 khz on either side of center is taken as a standard, but good results can be had from receivers which do not have this much bandwidth. If any portion of the curve needs improving, this can be done by tuning the signal genera tor to this frequency and adjusting alignment controls until the desired voltage output is reached.
One precaution should be emphasized in all alignment operations-do not overload the stages with too much input signal, as this distorts the response and results in over-correction with the alignment controls. When this has happened, the finished product will probably sound somewhat worse than it did before alignment was begun.
Fig. 12-20. A typical AM-FM receiver with ratio detector.
Fig. 12-20. A typical AM-FM receiver with ratio detector. (Cont'd)
AM-FM COMBINATIONS
Fig. 12-20 shows a typical AM-FM combination that includes all the features we have been discussing. The AM tuner consists of a 12BE6 in the mixer/ oscillator circuit with its output tuned to 455 khz. B+ to this stage is switched on by the AM-FM selector switch.
The windings of the AM and FM IF transformers are connected in series, with the AM transformers being left in the circuit while receiving FM, and vice versa. This is a very popular arrangement. It is possible to do this because the FM winding presents very low impedance at 455 khz, and the AM winding is connected on the "cold" side of the FM windings where the added impedance does affect the FM signal.
The AM detector uses the grid and cathode of V5, the FM limiter stage. For AM reception, plate and screen voltages to the stage are disconnected by the AM-FM selector switch, and rectification of the AM signal takes place using the grid and cathode as a diode. The FM detector, V6A, is a typical balanced ratio detector with the output between Test Point IT!] and ground. The AM-FM selector switch chooses the output from either of the detector circuits and sends it to the audio section.
When compared to Fig. 12-20, the receiver shown in Fig. 12-21 illustrates how much these AM-FM combination receivers can differ. The FM detector, V6, is a modified discriminator with the balanced output resistors, R22 and R23, connected differently. The output signal voltage is taken between pin 5 of the tube and ground. The packaged circuit, K4, is called the de-emphasis network. Its purpose is to reduce the response on the very high audio frequencies which are always emphasized at the transmitter in order to improve the signal to-noise ratio. Since noise is always greater on the high-frequency end of the audio spectrum, the high notes are trans mitted with greater-than-normal volume so that when they are reduced to normal in the receiver, the noise, which has not been emphasized, will be reduced to below normal.
The tuner incorporates an unusual feature by using one tri ode, V2A, of the ECC85/6AQ8 tube as a combination FM mixer, FM AFC, and AM oscillator. A study of the schematic will show that the AFC function is a typical reactance tube connected across the FM oscillator tank, L3.
The AFC control voltage is taken from Test Point ~ in the discriminator, filtered by R25 and C25, and finally connected to the grid of the AFC reactance tube. The same tube serves as a mixer on FM by combining the oscillator signal with the incoming RF in the plate circuit.
On AM, the tube serves as the oscillator by switching its cathode through L9 which is, in turn, coupled back to the plate through C14. The RF amplifier for AM is V3, the FM IF stage.
By switching the AM antenna directly into its grid through C16, and leaving the plate untuned for AM signals, the stage becomes an untuned RF amplifier coupled to the AM mixer, V4, through Cl8. The mixer grid is tuned by the tank circuit of L8, and the mixer is coupled to the oscillator by the common cathode connections in L9.
The AM IF amplification is accomplished by using 455- khz transformers in series with the FM transformers. The 6EQ7 is the last FM IF amplifier ( the limiter), and it also contains a diode plate (pin 8) which is used for AM detection. On FM, the stage is a limiter with the limiter grid-leak voltage (which can be measured at point <D> ) being used for A VC on the FM stages. While on AM, the stage is the second AM IF amplifier and detector combined, with A VC for the AM stages being taken at point <A>.
Fig. 12-21
TROUBLESHOOTING
Servicing straight FM receivers involves no more complications than those caused by more critical alignment. It will be recalled that distortion results from the slightest misadjustment of the secondary of the FM-detector transformer. While this adjustment is critical, the setup for it is not difficult.
All other failures fall under headings discussed in earlier sections, and the techniques explained there will suffice, with one exception-the problem of oscillator drift in FM receivers that do not use AFC is very difficult to handle. The usual measures are to replace the oscillator tube, followed by a systematic replacement of all resistors and capacitors in the oscillator circuit. When special temperature-compensated capacitors have been used in the original circuitry, they must be replaced with parts at least of equal quality and the same temperature characteristics. If no such capacitors have been used, perhaps the drifting can be reduced by putting some in.
When the AM-FM receiver must be serviced, the problems are greatly multiplied by the unorthodox circuitry and complicated switching networks which are used. A careful study of the schematic and the pictorial diagrams is necessary.
QUIZ
1. Compare AM and FM with respect to the following aspects; make a chart, and give comparisons in a few words.
FM AM
Change of power output with modulation Simplicity of receiver Simplicity of transmitter Effects of atmospheric noise Bandwidth required Limitations on audio frequency Broadcast-band frequencies assigned
2. Changes in the amplitude of the audio modulation applied to the FM transmitter cause what kind of changes in the transmitted signal?
3. Draw a circuit of a limiter stage giving correct values to all the parts. Include sketches of the waveforms at the input and output.
4. What determines the rate of change of the frequency of an FM signal?
5. Look up the schematic of an FM tuner using AFC and draw a partial schematic showing the AFC circuit. Ex-plain the operation of the AFC, and state the make and model represented.
6. In a small AM/FM combination, the audio is working but no signals can be received on either FM or AM. Make up a servicing chart showing a series of tests for locating faults that produce this symptom.
7. In what important ways do the alignment procedures for discriminators and ratio detectors differ? Explain with the use of schematics of the two circuits.
8. Explain why a slight mistuning of the detector input transformer secondary in an FM receiver causes distortion.
9. What is the effect of mistuning the primary of the detector input transformer?
10. Is an FM receiver always free of noise? Explain why or why not.