Radio Broadcast (May 1928-Apr 1929)

JUNE, 1928 FIELD STRENGTH MEASUREMENTS 103 Pt = Is2 Rt The radiated portion is given by: Pr = U2 Rr hs2 Rr = 1600 r-r Ohms A (4) (5) where Rt is the total transmitting antenna resistance, Rr is the radiation resistance, and hs and X are in the same units of length. Thus when we have determined the radiation resistance of the transmitting antenna by the procedure which is outlined by formulas 1 to 5, we have some idea of the energy radiated by the transmitting antenna, which means that we know the fundamental quantity in the physical functioning of the station when it is "on the air," We know how much of the power we put into it is getting away from the station. A practical little summary of work in field intensity determinations is the pamphlet by R. O. Cherry, prepared under the direction of Professor T. H. Laby, on "Signal Strength Measurements of 3LO, Melbourne." The measuring set consisted of a loop, a tuning condenser, and a thermionic voltmeter to measure the potential induced in the loop by the broadcasting station. The voltmeter was an instrument of the Moullin type, manufactured by the Cambridge Instrument Company, using plate rectification to cover a scale of 0-1.5 volts. Cherry's pamphlet does not show the voltmeter circuit, but this has been reproduced in Fig. 1 of the present description, from page 35 of Moullin's "The Theory and Practice oj HighFrequency Measurements" (Charles Griffin & Co., Ltd., London), an excellent work which has been reviewed in this magazine. Moullin's Fig. 25, added to Cherry's Fig. 1, gives us our Fig. 1. In the manufactured form of the instrument the plate battery is dispensed with and the 1.6-volt negative grid bias is secured from the 6-volt battery used to light the filament of the tube. Aside from this battery the voltmeter is self-contained. The calibration is stated to be independent of frequency and the galvanometer reads directly in volts, 1.5 volts r. m. s. being full-scale. When the applied potential difference exceeds 0.4 volt, grid current flows at the peak of the positive half cycle, and the instrument draws a slight amount of power, the apparent resistance at full scale being of the order of 0.75 megohms, corresponding to a power absorption of 2.5 microwatts. The voltmeter is used with British valves intended for a 4-5 volt filament potential, which is reduced to 3.5 volts in this case, thereby prolonging the life of the valve and the calibration of the voltmeter, barring accidents, almost indefinitely. The 3L0 report starts off with Formula (3) of the present discussion, followed by an expression for the field strength, whose equivalent is: E = (6) where E, in volts per meter, is the field strength at the point of reception; V, in volts, is the potential difference measured across the loop; hr, in meters, is the effective height of the loop w = 2xf, where f is the radiated frequency; L is the inductance of the loop in henrys; and R is the high-frequency resistance of the receiving circuit, at the frequency f. It is easy to see how, according to (6) the field strength, by definition, will equal the received voltage divided by the effective height of the receiver, but the origin of the square root factor may not be clear. Cherry gives no explanation, so it may be added that the expression without the added factor would be true for an open loop picking up a voltage from the transmitting station in question, but in practice it is necessary to tune the loop, as shown in Fig. 1, both in order to select the e. m. f. from the desired station, and in order to get enough voltage to measure. But then we are measuringthe resonance e. m. f. of the loop circuit, and this must be cor w2 L2 , rected before we can deduce the field strength. The next step is to measure or calculate the inductance of the loop. For the calculation process the reader is referred to the Bureau of Standards circular cited above. If a calibrated local oscillator is available, and the condenser across the loop also has a known calibration, the distributed capacity of the loop may be determined, and the inductance is then easily calculable from the wavelength formula: x = 1.885 x io» y~Cc (7) where C is the total capacity (loop capacity plus condenser capacity). Fig. 2 shows a curve of wavelength against various capacities of the tuning condenser when the loop circuit is tuned to different frequency settings of the oscillator coupled to it. The line being extrapolated, the point where it cuts the X-axis (zero wavelength) gives the dis DISTRIBUTED CAPACITY OF LOOP 2000 2 1000 65 100 200 300 400 500 600 700 800 CAPACITY MMFD. FIG. 2 tributed capacity of the loop. In the case of the 3 lo experiment the wavelength was 371 meters, and the loop capacity was found to be 65 micromicrofarads, which, added to the condenser capacity for resonance, gave a total capacity of 505 micro-microfarads, whereupon substitution in (7) gives: ■ 371 = 1.885 X 10V L X 505 x io " L = 7.698 x 10-5 Henry = 77 Microhenrys The resistance of the loop, condenser, and voltmeter circuit must also be measured. Cherry gives the procedure, but instead of repeating it I shall refer those who have a practical interest in the problem to Circular No. 74 again. The mean of several measurements in the particular example we are following was 3.5 ohms. A source of inaccuracy which must be considered at this point is that the resistance of the thermionic voltmeter is not quite constant, introducing a variable loss, which Cherry believes is between 0.5 and 0.9 phm equivalent series resistance. This will result in a slightly low value for the higher field strengths, but as such measurements, made with a loop, are not good to better than 5-10 per cent., a mean value for the high-frequency resistance of the receiving circuit is sufficient for practical purposes. All the other quantities needed for the calculation, first of the effective height of the loop (Formula 3), then of the field strength (Formula 6), are known. Using Formula 3, we may write: nr = 2 x <'-°3)*(5) 371 both the length of the side of the loop, which, squared, gives the area, and the wavelength, being expressed in meters; the result gives hr, also in meters, as 0.090. To get E from (6) we w L must find the value of w is 21c f, and f is the frequency corresponding to a wavelength of 371 meters; this may either be looked up in a wavelength-frequency table or calculated from the basic relationship that the frequency is the velocity of light (3x10s meters per second) divided by the wavelength, whence: and _ 3 X [O8 „ w = 2 1U = 5.08 X [O 371 w L (5.08) (io«) (77) (io-«) R 3-5 So finally we get E V (0.09) (1 12) IO.I This gives us the field strength in terms of the reading of the thermionic voltmeter, divided by a substantially constant factor, as long as the same loop, condenser, and voltmeter are used, and the wavelength remains the same. The loop, of course, is turned to secure a maximum deflection for each observation. If the apparatus were to be used for only one station, the wavelength of which is fixed, the voltmeter scale could be arranged to read the field strength directly. In the case of 3 lo, the following field strengths in m/v per meter were found, using the procedure outlined : Distance Direction 1 Mile 5 Miles 10 Miles North 400 90 45 East 250 60 30 West 200 50 From such data it is possible to draw contour maps of the field pattern of a station, such as those secured in the elaborate investigation of the distribution of weaf and wcap (Bown and Gillett: "Distribution of Radio Waves from Broadcasting Stations over City Districts," Proc. I. R. E., Vol. 12, No. 4, August, 1924). Bown and Gillett used one of the short-wave measuring sets developed by Bown, Englund, and Friis, and with this more elaborate apparatus were able to get down to field strengths of the order of a fraction of a millivolt per meter; some of their curves extend to a distance of over a hundred miles from the transmitter. The simple apparatus described by Cherry is, of course, restricted to a much smaller radius, but it illustrates the principles involved just as effectively. The contour lines in the case of 3 lo form a group of quite regular concentric ellipses, which would be expected with the transmitter located in fairly open country. The pattern from a station located, like the old weaf, in the heart of a city like New York, is far more irregular, naturally. Within the radius of neglectable absorption the product Ed (field strength times distance from transmitter) is approximately a constant; this relation may be used as a check on the accuracy of the field strength measurements. Goldsmith ("Reduction of Interference in Broadcast Reception," Proc. I. R. £., Vol. 14, No. 5 October, 1926) gives the following table of program service as a function of field strength: Signal Field Strength 0. 1 millivolt per meter 1. millivolts per meter 10. millivolts per meter 100. millivolts per meter 1000. millivolts per meter Nature of Service Poor service Fair service V ery good service Excellent service Extremely strong Edgar Felix has pointed out that the commercial value of a broadcast transmitter, other things being equal, is a function of field distribution. This is true, and, when the owners of stations realize it, more field strength measurements will be made in divers neighborhoods. If the field strength is not being produced in the sections where it is wanted, the artists and the advertisers might as well go home, and the studio be converted into a salesroom for artificial flowers. The limitations of space will not permit a longer technical discussion of the subject. Readers who are interested beyond this point are again referred to Moullin's book, in which Chapter VIII (pages 218-254) is devoted to a thorough study.