Hi-Fi AM Tuner and Amplifier

Part I: A Variable Bandwidth AM Tuner, by D. V. R. Drenner. This article first appeared in Radio Electronics, November, 1950

Most listeners still rely on standard AM broadcasts for the greater part of their radio entertainment. There is some pretty good fidelity in AM, if you have tuner and an amplifier to reproduce it Make the amplifier versatile by simply adding a selector switch and a bass control, and you can reproduce any hi-fi signal, whether AM, FM, or disc or tape recording.

This AM tuner doesn't depart from the conventional in too many respects. The superhet circuit, when it has variable-bandwidth i.f.'s, can pass all you'll get on the antenna, with little sideband cutting. Add a couple of other features and it will do a surprising job!


AM tuner

The schematic for this tuner (Fig. 1) shows some of these little additions which give it hi-fi at low cost - a feature most of us are interested in. For one thing, eliminating the conventional cathode bias on the r.f. and i.f. stages does away with numerous resistors and bypass capacitors. This makes a cleaner wiring job and reduces the over-all number of components. A bleeder voltage divider puts the correct voltage on plates and screens, confines the dissipated heat where it won't bother coils and impregnated parts, and adds to stability. To compensate for all the parts and money saved by this feature, separate decoupling networks are used in all plate and screen leads. But without them some serious common-impedance-coupling problems might arise, especially when the gain is rather high. The screens are run near their maximum voltage for just this reason- a compromise between high plate resistance and a high transconductance.

The diode detector was scrapped in favor of the infinite-impedance detector which is just as capable of handling large signals as the diode and which handles small signals with less distortion. The infinite-impedance detector is nothing but a cathode follower with a large cathode resistance to hias the tube almost to cutoff. This detector presents a very high impedance across the secondary of the last i.f. transformer so that there is no loss in Q as with a diode detector which has a relatively low impedance.

There we have the most important features of a good AM tuner. There are some others, like variable a.v.c., which we will mention.

The construction—whether it's on a new aluminum chassis, as the author's unit was, or on a well-baked cake pan— must be done carefully. The parts layout shown in the photos has short, direct leads, adequate separation of components and shielding, and gives stability in the r.f. and i.f. stages where it's really needed. And a hot r.f. stage can generate a lot of things beside a strictly class-A signal. The 6SK7 is used in the r.f. stage because it keeps the grid circuit where it belongs, down under the chassis with the antenna coil; and if the layout shown is fol lowed, there should be no unwanted coupling between circuits. The 2,200- ohm resistors and the two O.1uF capacitors in plate and screen leads are decoupling units.

The converter uses a 6SA7. This tube has a higher gain at broadcast frequencies than similar converters because of a fairly high conversion transconductance and a high plate resistance. Because of its construction it is reasonably stable under a.v.c. conditions, and with a separate oscillator its operation can be further improved.

A separate oscillator tube (a 6J5 in thiS case) reduces the chances of freuency drift due to the a.v.c. action on the converter. At broadcast frequencies this may be splitting hairs, but the theory says it can happen even there. So as a refinement worthy of a good AM tuner, a separate oscillator it is! As in the r.f. stage, separate decoupling networks are used in the plate and screen leads.

The i.f. stage has the older type 6K7. This puts the grid lead above the chasis and away from the plate lead, and reduces the chance of regeneration. Although the shielding of the single ended type tubes is good the 6K7 is better for the variable i.f. transformer used since the grid lead is on the top of the can. And here we have another chance at getting some fidelity. The secondary of the input i.f. has a tapped auiliary winding which broadens the selectivity curve and lessens the side-band cutting which otherwise would be rather severe when the i.f. is peaked. Other methods of doing this same thing use variable coupling between the two windings, but the result is the same-the top of the i.f. curve is broadened, and the skirts fall less sharply.

Following the i.f. stage is the 6C5 infinite-impedance detector. Since we need some a.v.c. voltage, which this detector can't supply, a separate tube, a 6H6, is used. This also provides a diode for the d.c. bias on the grids of the r.f., converter, and if, stages. One half of the 6H6 is led through a 50-uuF mica capacitor from the primary side of the i.f. transformer. This voltage will be higher than that delivered to the grid of the detector, and will give delayed a.v.c. This a.v.c. voltage is fed through the 1-megohm resistor in the a.v.c line, and then to the grids through their isolation resistors.

The R-C constant of the a.v.c. bus is made variable, along with the variability of the i.f. transformer. On position 1 of the bandwidth switch, the R-C constant is fairly short (0.24 sec.), so that the a.v.c. will follow normal moderately rapid fading. But as the bandwidth is increased (positions 2 and 3, the a.v.c time constant is progressively increased, giving 0.36 and 0.66 sec. at maximum. This is necessarily 1ong give good bass response: but once hi-fi broad position is normally used on local or other strong stations, this long R-C constant does not impair the a.v.c. action, such strong signals being usually fairly constant in level.

On strong signals the gain is reduced of course, and the long time constant takes appreciable time to return the gain to its normal high level. For this reason, tuning with the bandwidth switch in BROAD position will give an almost dead response if you tune from a strong signal across the band. until you hit another strong signal or until the a.v.c. voltage falls and increases the gain. It's a kind of unintentional quench circuit, but not bothersome since tuning is normally done in the SHARP position.

Along with the a.v.c. voltage, a fixed bias of about 3 volts is supplied from the other half of the 6H6, through the 0.1-uuf capacitor coupled to the 6.3 heater voltage. The two 100,000-ohm resistors and the 50-uF electrolytic form a filter and divider network for this bias voltage, which is fed through a l-megohm resistor to the a.v.c. hus. Thus, the proper bias is provided without the conventional cathode resistors, and there is no chance for the cathode to go more negative than the grids when a strong signal hits the grids.

When the tuner is completely wired, plug in the a.c. and connect a pair of phones to the audio output. The detector will provide a healthy signal for tuning purposes via phones if you haven't an amplifier handy.

If you've got a scope and sweep generator handy, aligning the i.f.'s is easy, and they'll have about 20-kc bandwidth in the BROAD position, and 10 kc or less in the SHARP position. Tracking and other alignment procedures are conventional.

Materials for Tuner

Resistors: 6-2200, -22000, -30,000, -47,000, 5-100,000 ohm, 1/2 watt; 2-1 megohm, 1/2 watt; 1-25,000ohm, 50 watt, wirewound adjustable; 1-1 megohm potentiometer.
Capacitors: 3-50 uuf, 2-.001 uf, mica; 2-.05, 4-0.1, -.25 uf, 200 volt, paper: lO-O.l uf, 600 volt paper; 2-16 uf, 450 volt l-50uf. 50 volt, electrolytic; -350 uuf trimmer; l-365 uuf, 3-gang tuning.
Transformers and coils: 1-antenna; 1-r.f. Interstage; 1-456 kc if, input with variable bandwidth; 1-456 kc if. output; 1-oscillator coil; 1-20-h, 70-ma choke; -350-0-350 vac., 70 ma, 5v at 2 amp, 6.3 at 3 amp power.
Miscellaneous:1-2-gang 3-position switch; 1-s.p.s.t. switch; tubes, sockets, chassis, fuse, fuse holder, dial, hookup wire, and assorted hardware.