Radio Broadcast (May 1928-Apr 1929)

Fig. 2 the emission curve of one filament and anode of this tube, shown by the dotted line, (Fig. 2) should show no evidence of saturation within the limits of the figure. In the oscillograph record, Fig. 3, each figure contains two separate records of the performance of the socket-power unit in question. The records made by vibrators Vj and Vi, labelled A in the figure, show in each case the performance under a load current of 20 mA. the normal maximum for this unit, while curves b show the performance at 50 mA. output. Referring to Fig. 3a, the upper record, Vibrator V3, is that of the voltage developed across the secondary of the transformer. Oscillograph Records THE voltage developed across the tube is V2 when the current indicated by Vi is flowing. It can be noted in the curves showing the current wave form that the voltage rises rather rapidly as the current starts to flow, and then less rapidly as it reaches higher current values. This is partly due to the fact that the resistance offered by the tube to the flow of current decreases as the current increases until the full emission current flows, and partly to the less rapid increase in instantaneous transformer voltage, near the peak of the wave. The current does not begin to flow at the instant the transformer voltage increases, but lags about 40°, this lag being due to the voltage existing across the first filter condenser. In this particular case the voltage across the load was 108 volts, and the voltage across the first filter condenser was somewhat higher. At the instant the current started to flow the transformer voltage was approximately 120 volts, indicating that this value of voltage existed across this condenser. Current then flows for the next 100° of this alternation, and at the time it ceases to flow the transformer voltage is 145 volts, as a result of the charging of the condensers. During the remaining part of the alternation the rectifier is idle. The voltage across the tube just reached the value required for full emission current, 50 volts, and the current was slightly higher than was expected from the d.c. readings, above 50 mA. On the second alternation the current reached a slightly higher value, 55 mA., but because of the higher emission given by tube, T13, the peak voltage drop was 43 volts. Conditions indicated by record b on these two vibrators were with maximum obtainable load current, 50 mA., to obtain which the output terminals were shorted except for the resistance of the vibrator. The voltage across the tubes reached a value close to the full transformer voltage, -Time -V 1-93.3 Volts/mm Fig. 6 the peak with tube Tu being 160 volts, and with T13 just under 150 volts. Practically all of the energy delivered by the transformer, 16 watts exclusive of the filament energy, was dissipated in the rectifier tubes, accounting for the fact that the output terminals had to be shorted to reach this current value. At the conclusion of this test a single cx-313 was substituted for the two cx-301a's with the following results. At 20 mA. the performance is practically identical with that obtained from the cx-301a tubes, but at 50 mA., the cx-313 has ample available emission so that the saturation current is not reached, the peak value rising to 115 mA. The voltage drop across the tube is much smaller as a result, the peak being 70 volts, as compared with 160 and 150 volts, respectively, with the cx-301a. tubes. Since less voltage is consumed (B) <D> 20 h. i Jj.TTT (C) <D> o T T (E) (E) Fie. 4 in the tube, more power is available in the output circuit and it is no longer necessary to short the output terminals to obtain this current. The output voltage, in fact, reached 60 volts, the output current being practically the same as before, 51 mA. As a result of this increased load voltage the voltage across the first filter condenser is increased, and the current no longer starts flowing as soon as the transformer voltage begins to rise, although it does carry current earlier in the cycle than was the case with the lower load current. The results are tabulated in Table I. The curves shown in Figs. 4a to 4e, record the performance of a half-wave rectifier, the type used being cx-381. The transformer potential, indicated by vibrator V3 on each record, was maintained throughout at 750 volts a.c. The load resistance was also kept constant at 5000 ohms. Half-Wave Rectifiers JN THE first oscillograph record, 4a, all filter elements are omitted. The peak current value was 140 mA., while a d.c. meter in the load circuit, indicating the average current, gave a reading of 47 mA. Thus, the tube's filament was called upon to supply, momentarily, three times the average load current. This is an important fact, since the filament must be made large enough to supply the emission current for the peak value. The ratio between peak and average current will be noted for each figure, in order to determine the relative load imposed by each circuit upon the tube's filament. The second exposure, Fig. 4b, was taken after adding a 4-mfd. condenser across the load. The charging of this condenser permitted a much larger current to flow through the tube, because of the reduced load impedance and the output current not only increased greatly in value to 102 mA., but continued to flow during the alternation when the rectifier was idle. The back voltage developed across this condenser reduced the time during which the tube carried current, thus further increasing the peak current demand upon the tube which, as indicated by V2, rose to the value of 535 mA. The ratio of peak to average current was thus increased to f§| or 5.2:1. In Fig. 4c, a 20-henry choke was placed in series with the load. The effect of the self inductance of the choke in causing the current to lag behind the voltage is quite evident on this film, and it also caused a marked decrease in the peak current to 70 mA., as well as in the average current (26 mA.). The current flowed for a longer portion of the cycle, however, and the ratio of peak tube current to average load current was reduced to the more favorable value of 2.8:1. A transient voltage was developed at the moment the current ceased to flow through the tube, caused by the self-inductance of the choke. In the next figure of the series, 4d, the usual filter system was added, the performance being similar to that obtained with a condenser alone, except for the improved filtering of the output current. The peak tube current was practically the same as in 4b, 540 mA., the output current 102 mA., a ratio of 5.3 to 1. In the fifth figure, 4e, the usual input condenser was omitted, resulting in greatly reduced demand on the tube and also in a marked reduction in output. The results are similar to those of Fig. 4c. except that the choke was smaller, 10 henries, and, therefore, had a smaller effect in reducing both peak and load current. The peak tube current was 130 mA., and output 45 mA., a ratio of 2.9:1. The transient voltage which appeared in Fig. 4 c was again present, and was sufficiently severe to result in an appreciable ripple in the output current. This series shows quite clearly that omission of the first filter condenser in a half-wave reo • april, 1929 page 394