Radio Broadcast (May 1927-Apr 1928)

210 RADIO BROADCAST JANUARY, 1928 show that the amplification is a factor of the mutual conductance of the tube and the external impedance. This external impedance is the effective resistance of the tuned circuit, as already mentioned, and varies as shown below for average coils at usual frequencies. The possible voltage amplification may be easily calculated, with an assumed mutual conductance of 400 micromhos and an infinite plate impedance. Frequency, Ro = L2or OHMS Amplification Kilocycles Rs Gm x Ro 100 400,000 200 1,000 100,000 40 10,000 10,000 4 These values of amplification are considerably greater than is possible with standard tubes such as we all use at the present time. At 1000 kc. (300 meters) the average gain in modern receivers may be as high as 10, and not many sets can do as well without some loss of stability. Actually, however, these values in the table with the 222 tube will not be attained, since the assumptions on which they were calculated, infinite plate impedance, no grid-plate capacity, and an effective resistance in the plate circuit of 100,000 ohms, are not realized. Since the tube's internal impedance is of the order of a half megohm, or greater, which is considerably more than can be attained by average coils or by coils which will not cut side bands, there is no use in FIG. 6 If a grounded shield is placed around the plate, the lines of force from grid to plate will be interrupted, and fewer changes of plate voltage will affect the grid — in other words, the grid is shielded from the plate above. If this resistance is equal to that of the tube, approximately one-half the mu of the tube may be realized. With 100,000 ohms in the plate circuit approximately one seventh of the amplification factor, or about 35 may be expected. For maximum voltage gain the effective resist amplifier, which had attached to its input a single wire antenna about 35 feet long, and a ground. With the brass box containing a Rice neutralized amplifier, using a 201 -a tube, and with the best position of the plate tap on the detector coil (see Fig. 8), the voltage upon the input to the detector was measured. Then the screened grid tube was used, the whole coil being used in its plate circuit, and the neutralization apparatus was done away with. The voltage was again measured with exactly the same input, and at the same frequency — 500 kc. In this case the output voltage was a little over three times the best that could be obtained with the 201-A circuit. Resonance curves showed that the two circuits were about equally selective. If the 201-A tube gave an amplification of ten, which is reasonable, the new tube had an indicated voltage gain of over 30, which seems to fit in with our calculations explained previously. Two stages would give a gain of 900 compared with 100 for two 201-A amplifiers, or approximately 20 tu, which has about the same effect as adding one stage of audio to existing receivers. With the antenna described, and with but two tuning circuits, there was no difficulty in separating weaf and wjz, 50 kc. apart, when the former was 8 miles away with 50 kilowatts of power in the antenna and the latter was roughly 30 miles distant with somewhat less power. Measurements showed that, with the screened grid tube, weaf delivered over 4.5 volts to the detector, sadly overloading it. 201-A Oscillator ? FIG. 7 In the Laboratory this arrangement of apparatus was used to discover the voltage delivered to a detector when the screened grid tube was used as an amplifier. The meter in the detector plate circuit read the change in plate current when a. c. voltages were applied to its grid. It was a calibrated detector, or vacuum-tube, voltmeter To Osc. Here the vacuum-tube voltmeter was applied to the output of a standard Rice neutralized amplifier. The difference between the input voltage in this case and that in Fig. 7 determines the gain in using the screened tube. The meter used in the calibrated detector was a Westinghouse 200microampere meter, type px using a step-down transformer to couple the output of the tube to a succeeding stage, and the whole coil may be used without danger from oscillation. With the entire coil in the plate circuit, as shown in Fig. 7, the amplification per stage may be figured by somewhat simpler mathematics. In this case we have a simple tuned impedance in the plate circuit of the tube, which at resonance has an effective resistance, as explained CX322 UX-301-A »180 0.006 FIG. 9 When the grids of the ux-222 (cx-322) are reversed, i.e., the inner grid is made positive, the amplification factor and plate impedance fall to more usual values, and the tube can be used as an audio or intermediatefrequency amplifier. ance muif be high, that is, we must use exceptionally good coils of high inductance and low resistance with the probability that fidelity will suffer. In the Laboratory a simple set-up was employed to examine the tube's behavior. The circuit is shown in Fig. 7 and, as may oe seen, consisted of a single amplifier tube followed by a non-regenerative C bias detector which could be calibrated in volts input against change in d. c. plate current. The shielded tube and its accessory input apparatus was carefully enclosed in a tight brass bo:, which was grounded. The audio amplifier was useful as a kind of monitoring stage to follow what was happening in the preceding circuits. It made it possible to insureagainst theamplifier oscillating, etc. A 1000cycle modulated signal was induced into the The tube may also be used as a space charge grid affair, where the inner grid is positive with respect to the filament and the fine structure which ordinarily screens the plate is used as the signaf grid. To test its capabilities the circuit shown in Fig. 9 was used. With an input voltage of o. 1, the gain was about 30, from 60 to 30,000 cycles, and then fell slowly. At the higher frequencies the tube capacities (in the space charge grid tube the capacities are much greater than in the screened grid connection) shunt the output resistance, with consequent loss of amplication. Here, then, is a tube which must henceforth be carefully considered by all designers of radio equipment, for amplification of the very high frequencies, those of intermediate range such as are used in super-heterodyne receivers and broadcast frequencies, and again at the low audible tones. In all of these ranges it seems probable that greater amplification will be secured than has been possible with ordinary or high-mu tubes. In subsequent articles, we shall discuss the design problems at greater detail and have something to say about the problem of selectivity and fidelity of reproduction.