Radio Broadcast (May 1928-Apr 1929)

Simplified Autodyne Circuit Used in A DOUBLE-DETECTION SHORT-WAVE SET By ROBERT S. KRUSE View of the converted short-wave tuner, i.f. amplifier, and a.f. amplifier. IN THE writer's article published in February Radio Broadcast the advantages of double-detection receivers over other short-wave receivers were discussed. In order to facilitate the comparison there was described a species of adapter which may be applied readily to an ordinary detector-audio set, converting it into a doubledetection (super-heterodyne) receiver. Since the device employed a heterodyne oscillator two tuning controls were required. This same complication also existed in the two other forms of the circuit which were described, namely a double-detection adapter (to proceed an ordinary broadcast receiver) and an outright short-wave receiver of the doubledetection type. When it is desired to simplify the control and retain the advantages of the tuner, one naturally thinks of combining the tuning controls. The first suggestion is that this may be accomplished by the use of a two-gang condenser after the universal practice employed in broadcast-receiver construction. The solution is not satisfactory, however, since the problem is not the same as the one encountered in the 500-1500-kc. band. The broadcast designer or builder has to make only two coil-condenser combinations work together, but in short-wave work we would be required to make the circuits remain in alignment with four or five sets of coils which are plugged into tuner and heterodyne, respectively. Of course, this can be done, but commercial coils are not matched accurately enough for the purpose since the makers have not anticipated such an arrangement. Indeed, it would be difficult to make them sufficiently alike at a cost approaching the coils now on the market. One may then leave this idea and turn to the alternative, which is to avoid the necessity of tuning two circuits by the process of omitting one of the circuits; patently this involves a transfer of the second circuit's functions to the remaining circuit which must now serve two purposes. Therefore, we may proceed by investigating the possibility of combining the functions of the oscillator and the first detector in a single tube and a single tuned circuit. The possible difficulties are loss of sensitivity, selectivity and audio quality. The audio quality consideration may be determined more easily by trial than by other means and the reader is asked to accept the rather dogmatic statement that in the arrangement which follows audio quality does not suffer. The selectivity is certainly not improved by the combination of the two circuits, but as it happens the present arrangement is one in which the i.f. amplifier supplies the selectivity and we are not so seriously concerned with that matter. Sensitivity of System WHEN the question of sensitivity arises one must confess that a definite loss has taken place by reason of the choice of 95 kc. as an intermediate frequency. However, this loss is not serious as the presence of a pair of screen-grid tubes in the complete system produces an overall gain that is materially above that of the system described last month, and is, in fact, above normal requirements. The choice of 95 kc. as an intermediate frequency is due to the desire to avoid any damage to audio quality, while at the same time avoiding an excessive amount of detuning of the autodyne detector in process of transferring the signal into the i.f. system. This contrary pair of considerations may require a word of explanation. If audio quality were the only consideration we would choose an intermediate frequency in the vicinity of, perhaps, 1000 kc, thus securing a noiseless amplifier and complete certainty that the harmonics of the oscillating detector would do no damage. This plan was followed in the February article with a separate oscillator (heterodyne). But with an autodyne (combined oscillator and first detector) we cannot use as high an intermediate frequency for we would then be compelled at all times to tune 1000 kc. off the desired signal so as to transfer it to the 1000 kc. amplifier. Such mistuning would, of course, weaken the signal materially whereas the detuning necessary to produce a 100 kc. beat is not fatal. Fortunately this — like the other difficulties — turns out to be an academic, and not a practical, difficulty. The i.f. system used consists of a pair of Rusco 95 kc. air-core transformers and a Rusco band-pass filter working at 95 kc. There are two ways of making a short-wave super -heterodyne, as Mr. Kruse pointed out in February Radio Broadcast. One involves turning a short-wave tuner into the frequency changer and one's broadcast receiver into an intermediate-frequency amplifier. In this article he tells how to make a receiver that starts with the antenna and ends with the audio output — and it is a double-detection set of considerable amplification and selectivity. It does not involve playing tricks on one's short or broadcastwave receiver. — The Editor The tuner with which the device has been associated in the writer's experiments is made by the National Company and has a wavelength range of 14.5 to 115 meters. At the 115-mecer end of the range a 95 kc. beatnote requires nearly 4 per cent, mistuning, which seems rather bad to one accustomed to broadcast work. At the 14.5-meter end the mistuning is about \ per cent. Fortunately one is saved by the very thing that suggested the band-pass, namely, the comparative lack of selectivity of a lonesome tuned circuit. In practice the signal obtained is not materially weaker than that obtained with a heterodyne, the rest of the equipment remaining the same. This is, to a considerable degree, accounted for by the fact that the strength of the oscillation was adjusted in all cases to a favorable value by use of the normal controls of the tuner, operating in the normal manner. Badiation from the autodyne's first detector is prevented by the 222-type tube in the first socket of the receiver. The circuits, which are shown in Figs. 1, 2, and 3, do not seem to require much explanation. However, some readers may be confused by the band-pass filter, but its purpose may be explained by the simple statement that its business is to pass only the band of frequencies lying between 90 kc. and 100 kc, while stopping lower and higher frequencies. It follows that the only signals to get through the system are those which the autodyne system has transferred into the "pass-band." The purpose of this device is, therefore, to provide the selectivity of the system and to suppress noise as well. Since the Rusco bandpass filter consists of four shunt sections (and the corresponding series parts), it is sufficiently complex to give a good flat top and sharp cut-off, unlike the usual arrays of tuned circuits. Adjustment of Filter ONE difficulty may arise which has caused several filters to be denounced as "no good." A filter, unless terminating in the proper sort of a load, will produce all sorts of « march, 1929 page 309 •