Radio Broadcast (May 1928-Apr 1929)

RADIO BROADCAST cycles oiF resonance, the input must be about 10 microvolts, or five times as much as at resonance. In Fig. 4 is shown the field strength pattern at a particular antenna, where the heights of the lines represent the amplitudes of the "carriers" on each channel frequency. The wavy shaded portion represents the "static level," or the voltages introduced in the antenna due to static and unwanted noise. This graph also contains the curve of Fig. 3 (curve a). The stations heard will be the one at 1000 kc, at which the set is tuned, and also the station on the adjacent channel, 1010 kc. The relative strengths of the signals will be about in proportion to the height of the fieldstrength lines above the selectivity curve. Thus the station on 1000 kc. will be about 4 times as loud as that at 1010 kc. The field strength of the station at 980 kc. is even greater than that at 1000 kc, but due to the selectivity of the circuit it cannot even be heard. The station at 990 kc. could never be heard with this set even if it were tuned right on 990 kc. A more sensitive set would be necessary, and even then the reception might not be good, due to the static and interfering signals. Now assume that the receiver used had poorer selectivity as shown in Fig. 5. (Hereafter only the superimposed diagrams are given.) It is apparent that three stations are heard at once quite strongly and a fourth, at 1020 kilocycles, can just barely be heard. A simple way to visualize this is to imagine the selectivity curve, such as in Fig. 3, is cut down into a sheet of cardboard, and then laid over top of the field strength picture. Sliding this cardboard horizontally along the frequency scale corresponds to tuning the receiver, and the stations you can see projecting up above the slot in the cardboard are those which you will hear. If you cut out such a cardboard cover having the selectivity curve shown in Fig. 1, and slide it back and forth as suggested, you will see that it will be impossible to get any one station without interference from some other station. This is too often true with many radio receivers. Effect of Tuning A VIEW of the selectivity field strength charts will show a well-known fact. Detuning the receiver will reduce the output, and therefore, this method is often used to control the volume of a set. There are several objections to this practice. The resonance response to some interfering frequency is increased greatly, thus decreasing the selectivity and increasing the noise. In addition this detuning invariably changes the quality of reception. The receiver should be tuned directly on the station desired and the output regulated with the volume control. There are many other peculiar cases of selectivity, some of which are illustrated here. IB LOW FREQUENCY RESONANCE H,GH frequency Fig. 2 — Selectivity is improved by connecting two or more resonant circuits in series. Curve A shows the selectivity of a single circuit, curve ? of a series of circuits Take as an example the case of a strong local station. Such a station may have a field strength of as much as 10,000 microvolts, per meter, which is far beyond the range of the cross-section sheets used here. For illustration, take the curve of Fig. 1. Notice that the horizontal scale of frequencies is smaller than before. The selectivity is the same as in Fig. 5. Assuming we are using the cardboard covers again, they must be placed in the positions shown in order to avoid hearing the strong local station. This means that no station between 650 and 750 kc. can be heard without interference from the local. It may be that before the local station raised its power, the radio set could satisfactorily receive the station on 720 kc. Therefore, when such reception can no longer be obtained, the local transmitter is blamed. The transmitter is "broad" and it "spreads all over the dial" and many are the protests against the design. However, the only crime of which the station can be accused rightfully is that of attempting to increase its service area and providing the benefits of static-free reception to a greater number. Investigation of such complaints has shown almost always that they have originated with the owners of obsolete receivers, at least from the selectivity standpoint. As bigger and better transmitters are built, receiver design must necessarily conform. It is unfortunate that advancement in radio service must render obsolete the inefficient type of receiver and act to the 200 II I I I I I I I 950 960 970 980 990 1000 1010 1020 1030 1040 1050 FREQUENCY IN KILOCYCLES Fig. 3 — The selectivity characteristic of a receiver is usually drawn in the manner indicated in the above graph detriment of its owners, but it's the same old story of serving the majority. Having thus shown that the lack of selectivity is inherent in the receiver, a closer study of receiver selectivity is justified. And it is at this point that the meaning of the subject of this paper will become apparent. To an engineer, selectivity is represented by the curve shape, and the ratio of the signal field strength needed at any frequency to that needed at resonance to produce an audible signal. This is a very real thing, expressible in numbers, and is a factor which can be calculated as a ratio. To the user, the selectivity of a receiver is apparently the way in which stations can be separated. The relation can best be shown by the use of our cardboard covers and a field-strength diagram. This is shown in Fig. 6. The selectivity curves of two sets, a and b, are drawn. The real selectivity of the two sets is exactly the same. For simplicity, the field strengths of five adjacent stations are shown having equal amplitude. By sliding the cover marked a, it is obvious that any one station, to the exclusion of all others, may be selected. By sliding b along the frequency scale, it is apparent that at least, three stations will • april, 1929 . . . page 400 • FREQUENCY IN KILOCYCLES Fig. 4 — In this graph the selectivity curve (Fig. 3) is superimposed over the field-strength pattern at a particular antenna. Vertical lines indicate amplitudes of carriers and shaded area indicates the "static level" be heard at all times! The difference between ■ real selectivity and apparent selectivity is quite clear! Selectivity vs. Sensitivity PROPERLY understood, this fact is no detriment. By backing off the volume control on the more sensitive set until the two sets have equal sensitivity, the apparent selectivity will be equal. In general it may be said that the greater the sensitivity of a radio receiver, the less will be the apparent selectivity. (Notice that this does not say that the set having the least apparent selectivity will be the most sensitive.) In demonstrating a very sensitive receiver, a salesman will often turn the volume control full on, showing how many stations can be received. While not saying "Yes, all at once," the prospective customer may react against the set due to the apparent lack of selectivity. It is so simple to reduce the volume control and thus increase the apparent selectivity that all this discussion seems out of place. Yet so many operators of good radio receivers will unhesitatingly blame the poor selectivity of his receiver upon conditions entirely irrevelant, when a proper consideration of the true factors will clear up a lot of trouble. Ask any radio dealer how much trouble has been caused and how many radio sets have "suddenly gone wrong" due to a change in power, frequency, or location of the customary vendor of entertainment. Before leaving this point, it is well to drop a word of warning to any prospective purchasers. In the demonstrations of selectivity make sure the fine apparent selectivity of a set is due to the real selectivity of the circuits and not due to simply a lack of sensitivity. The selectivity of a receiver depends on the size of the coils and condensers, the resistance, circuit coupling, and many other circuit factors. As the set is tuned from one end of the broadcast-frequency range by changing the condensers or the inductance of the coils, the real selectivity changes. Fig. 7 shows three selectivity curves taken on the same receiver. (The sensitivity in each case was adjusted to interference output with two microvolts per meter input.) The selectivity at 550 kc. is quite good, while that at 1500 kc. is poor. This means that instead of having a constant selectivity, and thus a cardboard shield of constant shape, the gap must open up gradually as the set is tuned to a high frequency. As the usual receiver varies greatly in