Radio Broadcast (May 1927-Apr 1928)

224 RADIO BROADCAST JANUARY, 1928 one. Fig. 4 shows four typical calibration curves made with different values of plate potential. After one has become familiar with the vacuumtube voltmeter described here, the voltages may be adjusted and all kinds of different curves obtained. For the process of calibration, some known source of a.c. is necessary. A fairly accurate calibration may be made by using one of the many step-down transformers that are now being used to supply filament voltage to a.c. tubes. With the primary connected to the no-volt a.c. house supply the secondary voltage from one of these units will give several points on a perfectly good calibration curve. For example, a transformer which supplies filament power to a receiver using 226, 227, and a power tube, has the following voltage taps available: 5, 2.5, 1.5, 0.75. These voltages can be added to or subtracted from each other by connecting the windings to aid or to buck each other. Fig. 5 shows the complete vacuum-tube voltmeter with the accessory equipment for calibrating it. If, during calibration, but before the maximum known voltage to be applied across X and Y has actually been applied, the microammeter needle is dangerously near the maximum deflection point, it will be necessary to increase the C bias and commence the calibration again. Vacuum-tube voltmeters can be calibrated by any Laboratory at small cost and it is probable that the constructor can procure such a calibration near-by. The reader should remember that the tube which is to be used should be sent with the instrument and that the whole unit should be very well packed. The calibration of the instrument is practically independent of frequency, so that the experimenter can compare voltages in circuits operating at audio, intermediate, or broadcast frequencies. He can measure the voltage step-up in an audio transformer or amplifier, or the output of two radio-frequency amplifiers, or plot the resonance curve of an intermediate frequency stage. It will be noted that the input of the tube looks directly into the device whose terminal voltage is being measured. If d.c. flows through this device, it will be impressed across the input to the voltmeter and ruin either the calibration or the plate meter, or both. Some means, such as that shown in Fig. 6, must be provided for isolating the voltmeter from d.c. potentials existing in the circuit under measurement, when, for example, the output voltage of a resistancecoupled amplifier tube is being measured. With this arrangement the short-circuiting switch shown in the photograph on page 221 may be omitted, since the grid is always at a safe d.c. potential with respect to the filament. Frequencies in the same range, that is, all audio tones, or all frequencies in the broadcast band, will give similar calibration curves. If the experimenter has no means of calibrating his instrument, he is not so unfortunate as might be supposed. He may make use of the fact that the deflections of the microammeter are roughly proportional to the input a.c. voltages squared. Thus he may get two deflections representing the gain, say, of two amplifiers. He need not know the actual knows roughly how much g to the other. All he needs vo-H I I vAAAAAAi A suitable set-up for transformer which is taps on the secondary and these voltages provided he reater one is compared to do is to divide one 0.006 mfd. FIG. 6 When the vacuum-tube voltmeter is used to measure an a.c. voltage across the terminals of which is also a direct-current voltage, some means must be employed to prevent this d.c. from getting on the grid of the tube. Condensers as shown here will accomplish this. The diagram shows measurements being made on a resistancecoupled amplifier u.-Amps FIG. 5 calibrating the vacuum-tube voltmeter. T is a used with a.c. tubes and, therefore, has several . Thus several known voltages may be obtained, plotted as a graph, similar to Fig. 4 deflection by the other and extract the square root. This will be the ratio of the voltages. The home constructor should be warned again to carefully watch every step when experimenting with the vacuum-tube meter and to be very careful of the microammeter. One mistake and the meter is gone. Mways, at the conclusion of an experiment, remove the meter from the circuit first. Those who wish to read further about the vacuum-tube voltmeter will find a most interesting and instructive series of articles in the English paper, Wireless Experimenter and Wireless Engineer, for October and November, 1926. Part of this material was republished in this country in Lefax leaflets in February, 1927, and forms the best background for the serious student of this instrument. It is possible to purchase from the Cambridge Instrument Company, of Ossining-on-Hudson, New York, a vacuumtube voltmeter calibrated and "ready to go." It is a beautiful instrument, although somewhat expensive. The owner, however, may rest assured that he has the best possible apparatus for measuring small a.c. voltages, radio-frequency currents, the gain of his amplifiers, both audio and radio, and for the performance of a hundred valuable experiments. The vacuum-tube voltmeter, although an invaluable asset to the Radio Broadcast Laboratory, by no means holds a monopoly in usefulness. Here are just a few of the dozens of meters always in use in the laboratory. Included are instruments by Weston, Sterling, Jewell, Dongan, and Hoyt