Radio broadcast .. (1922-30)

.RADIO BROADCAST. Fig. 6 practical case, therefore, there will be no effect due to the secondary electrons aside from the heating of the plate, as long as the plate is at the highest potential. And this is actually the case in the usual uses of the tube. In measurements where a higher voltage is impressed on the grid to carry the characteristics be- yond the usual range, it is found that after the saturation current is reached, the emission is no longer constant but greatly in- creases with higher positive plate voltages. In this case, the elec- trons wander from the plate to the grkl and, as a result, are in the opposite direction to that of the normal emission of the filament. They therefore subtract from it and give the apparent effect that the plate current decreases. The secondary emission comes into the question, then, whenever there is another electrode in the tube which has a positive voltage applied to it. In the double-grid tube, due to the external voltage drop, the plate frequently has a lower voltage than the auxiliary grid. In this case, the secondary electrons will travel to the point of higher potential— that is, to the auxiliary grid—and as a result, in the upper part of the emission curve the functioning of the tube will be affected. A means must be devised to take care of these secondary electrons when the plate voltage is less than the auxiliary grid volt- age. This can be done by the insertion of a third grid, the so-called cathode grid. This is put in between the plate and the auxiliary grid and has a somewhat lower potential. Therefore, the secondary elec- trons find themselves in a field directed towards the plate, since they prefer to go to the plate rather than to the cathode grid with its lower potential. As a result, the secondary electrons fall back into the plate and do not affect the operation. A very interesting feature of the three- grid tube is the apparent improvement of reproduction at the higher frequencies. The high internal resistance, about 50,000 ohms, can be neglected in comparison with the resistance of the loud speaker at the higher frequencies. The loud speaker, of course, has a dropping frequency charac- teristic at the upper end of the audio range. The three-grid tube, due to its higher internal resistance, is the only tube which tends to give better reproduction of the higher frequencies and therefore a more even result over the whole range. Practicall y this is of no advantage, since usually the bass is preferred in a loud speaker, as is the case in an electrodynamic type. In the use of electrodynamic loud speakers, however, the three-grid tube has an ad- vantage due to the more natural reproduc- tion. [It must be remembered that the author is speaking from the Continental view- point when he discusses fidelity of re- Eo'z AUXILIARY GRID VOLTAGE -10 0 +10 GRID VOLTAGE + 20 50 100 SCREEN-GRID VOLTAGE Fig. 8 150 Fig. 10 production. Of course, it is possible to "match" any tube to any load with the proper transformer and thereby secure the desired characteristics.— The Editor] Measurements In the following paragraphs are the results of our measurements on a three- grid tube of the type described. A tube was connected as it is usually used in present loud speaker practice (See Fig. 3). The internal construction of this tube is shown in Figs. 6 and 7. Fig. 4 shows the relation between fila- ment current and emission for a grid voltage E s = 0 and E g = —15, as well as (i r ) between filament current and screen- grid voltage (i,j). It will be noted that the value of the screen-grid current is about 16 per cent, of the total load current from the plate current source. These curves correspond to the normal temperature Fig. 7 emission curves; the screen grid can be considered as a part of the plate. Fig. 5 shows the relation between plate voltage and emission to the screen grid and plate for two differ- ent grid voltages. In this case also, the screen grid takes about 16 per cent, of the load current. The mea- surements were carried only as high as 150 volts and at this voltage the saturation point had not yet been reached. Filament temperature was constant in these measure- ments, and the auxiliary grid was at the same potential as the plate. In the measurements of Fig. 8, the auxiliary grid voltage was varied and the plate voltage held constant at 135 volts. The emission and auxiliary grid current increase rapidly with increase of the auxili- ary grid voltage. This voltage should therefore be as high as pos- sible, even higher than the plate voltage itself. Even the slope of the characteristic depends to a large extent on the auxiliary grid voltage. It increases, as shown in Fig. 9, with an increase in the auxiliary grid voltage. The plate voltage was held constant at 135. The curve in Fig. 10 was derived on the assumption of a normal characteristic for the different valuesof the auxiliary grid voltage. It can be seen that not only the emission but the steepness decreases with decreasing plate voltages. 50 100 SCREEN-GRID VOLTAGE Fig. 9 150 • DECEMBER 1929 • • 117