Radio Broadcast (May 1928-Apr 1929)

RADIO BROADCAST 30 FREQUENCY IN KILOCYCLES Fig. 5 — The selectivity curve of Fig. 2A superimposed over the fieldstrength pattern of Fig. 4. Note that the signals of four stations are heard at once due to the poor selectivity of the circuit sensitivity in different parts of the frequency range, and as the real selectivity (as per Fig. 7) also varies greatly, the apparent selectivity is quite a complicated result. This elementary discussion should help considerably, however, in understanding the action of the receiver. Other Considerations THE real and apparent selectivity so far have been considered only on the electrical basis on which they operate. To the user of a particular set, there are mechanical factors contributing to the apparent selectivity almost as important as the electrical characteristics. A large frequency-control knob tends to make tuning easier and increases the apparent selectivity. The vernier drive and the large illuminated drum dials are contributors to the same idea. To the operator, who cannot "see" the field strength pattern at his antenna, the selectivity is estimated by the dial movement necessary to eliminate any particular station. This basis for judging selectivity is as inaccurate as it is common. The "results" obtained depend upon the frequency range covered by the entire dial movement and the rate of frequency change at any operating point. Assume two sets, both having a dial movement of half way around, one tuning from 540 kc. to 1520 kc, and the other tuning from 570 kc. to 1350 kc. It is evident that the apparent selectivity would be greater on the first set as a given movement would shift the resonance frequency faster. The shape of the condenser blades greatly influences the apparent selectivity. If the blades are semi-circular, a given angular rotation will cause a greater change in resonance frequency at the high frequencies than at the low frequencies. This results (as based on dial movement) in the apparent selectivity being greater at high frequencies, and thereby (to the user's mind) compensates the decrease in real selectivity as indicated in Fig. 7. If the condenser blades are shaped so as to give uniform change in frequency with uniform rotation (so called straight-linefrequency type) the apparent selectivity will be the same as the real selectivity. One more factor is responsible for a huge misunderstanding in judging the sensitivity of a radio receiver. That is distribution of the broadcasting stations themselves. The new allocation has placed a great number of small stations on the same, or closely adjacent bands. A sensitive receiver will pick up a great many of these stations, and owing to their proximity, several stations may be heard at once. Such stations are usually so close together that they interfere or "beat" with each other and produce a strong continuous audio note. This note is not due to the modulation of either station, but is the "beat" of the two carriers. Such interference is quite common and is usually understood by those who have a technical knowledge of the reasons why, but to a novice or to one who doesn't care to know why, such interference is charged against the receiver's selectivity. Most radio users know that the selectivity of a receiver enables them to eliminate those signals they do not want, and they automatically class those cases of station interference against the receiver also. Another type of station interference which occurs in some receivers is that obtained due to modulation in the receiver. Two signals may be heard at once, but only when tuned to either one of the stations. The signals of the two stations "modulate" each other in the r.f. end of the set and both modulation frequencies are superimposed on each carrier frequency. Obviously "selectivity" can do nothing to help such cases. Fortunately such cases are rare as they occur only when the volume control is in the audio end of the set, or where extremely strong signals are impressed on an untuned input. This kind of "cross talk" can best be eliminated by using a wave trap tuned to one of the two stations interfering. This may be an important point in selectivity 500 FREQUENCY IN 'KILOCYCLES ' Fig. 6 — The real selectivity of the two curves above is exactly the same, but the apparent selectivity of A is much greater demonstrations and the real selectivity impaired because of this fact. Conclusions WHAT conclusions of value to the set user may be drawn from the above discussion? First, and probably the most important, is the fact that the apparent selectivity can be increased greatly by decreasing the sensitivity of the receiver, and then tuning accurately on to the frequency of the station desired. Try it. You will be surprised at the big improvement in selectivity when the volume control is reduced just enough to eliminate the undesired station. Another valuable conclusion is in regard to the size of the antenna. With higher-powered transmitters rapidly coming on the air and "interference" reports on the increase, it is standard practice for the radio editors and engineers to say "Use a smaller antenna." It is obvious that this accomplishes the result. In Fig. 6, the apparent selectivity of set b will increase just as much by reducing the input to 7 microvolts, as it will by changing the sensitivity to correspond to the set a. However, reducing the size of the antenna to improve the selectivity decreases the distancegetting ability of the set. As previously shown, intelligent use of the volume control gives the same control of the apparent selectivity without permanently reducing the efficiency of the set. Here's one more vote for longer and higher antennas. On clear, cold nights you can use • april, 1929 . . . page 401 • the extra sensitivity a little brain work will provide. How selective should a set be? An ideal set should have sharp cut-off and a flat response the width of one broadcast band. To date no commercial set of this type has been built, and the problems involved indicate that such a receiver will not be available for some time to come. Just remember, in demanding "knifeedge selectivity," that, like lots of other things in life, selectivity is a compromise. Too broad a selectivity curve, and interference results. Too sharp a selectivity curve and the quality is impaired greatly by the loss of the high notes. A good engineering compromise, plus a little intelligent operation, will provide for many hours of interference-free entertainment. WARD LEONARD'S AUTOMATIC VOLTAGE REGULATOR UNSATISFACTORY receiver performance, and short tube life due to excessive fine voltage — this has been one of the most pressing problems associated with the a.c. receiver and has probably been one of the major causes of consumer dissatisfaction with sets of this type. It appears, however, that the problem has now been solved completely and in a very satisfactory manner by' the Ward Leonard Company. This company has perfected a new device (that many manufacturers will probably include in their receivers this fall) which performs the double function of power transformer and linevoltage regulator. This device makes the operation of the set independent of ordinary variations in line voltage. It takes the place of the power transformer ordinarily used in a receiver and it functions, not only to supply all the voltages required for the set's operation, but also to compensate variations in line voltage. With this new device in the set the line voltage can vary from say 90 to 150 volts and the actual voltages applied to the tubes in the set will vary a negligible amount. We were present when the Ward Leonard Company demonstrated this device which was installed in a Crosley set. By means of an auto transformer connected to the a.c. line, the engineers varied the line potential from 90 to 150 volts. The voltage applied to the filaments of the tubes varied less than a tenth of a volt! The device works entirely on magnetic principles — it contains no moving parts, resistors, thermal units, etc. The Ward Leonard engineer responsible for the design is H. K. Kouyoumjian. H. E. R. -20 -10 RESONANCE +10 +20 KILOCYCLES Fig. 7 — The selectivity of the usual receiver varies with the frequency to which the circuits are tuned as illustrated in the above curves