Radio Broadcast (May 1927-Apr 1928)

222 RADIO BROADCAST JANUARY, 1928 Builders of plate-supply units, and owners of power amplifiers, who have placed a milliammeter in the plate circuit of the power tube, and watched its deflections under strong signals, have possessed the essential part of a vacuum-tube voltmeter — an overloaded tube. Whenever a signal came along which was greater than the C bias could handle, the average plate current changed and caused variations of the milliammeter needle. But why does the needle wobble? The normal plate current, as indicated by the milliammeter, is fixed with a fixed value of grid bias and a given plate voltage. When small a.c. voltages (the incoming signals) are placed on the grid circuit, the C bias is changed accordingly and in the plate circuit appears a magnified replica of these input voltages. In other words, the actual voltage on the plate of the tube varies with the input grid-filament voltage and naturally the plate current changes accordingly. If these changes are rapid and symmetrical, so that an increase in current is followed by an equal decrease, the average value of the plate current will remain the same and the milliammeter needle does not move. It is these a.c. plate currents which are amplified and which produce signals: the d.c. current is only a necessary and not directly useful part of the process. If the changes in plate current are not symmetrical with respect to the value when no a.c. input is applied, the average value of plate ■current is different, and the plate milliammeter needle jumps about if this average value changes rapidly. The relative amount of plate current change, and whether it decreases or increases from the steady d.c. value with no input, depends upon the fixed C bias, so that we can make the complete unit, tube and meter, into a sensitive indicator of small a.c. voltages if we choose the plate and grid voltages correctly. XO YO n-Amps FIG. 2 Here is the circuit diagram of the vacuum-tube voltmeter described in this article. The additional apparatus not shown in Fig. 1 constitutes a bucking-voltage device to keep the normal plate current out of the microammeter and, of course, the filament battery and rheostat The vacuum-tube voltmeter, then, consists of a tube so biased that input a.c. voltages cause large changes in average d.c. plate current. To make it into a calibrated meter, one merely places known a.c. voltages on the input, reads the plate current change, and plots a curve (typical curves are shown in Fig. 4). The purpose of the bypass condenser, Q, shown in Fig. I, is to improve the rectifying or distorting property of the tube. To prevent the device taking power from the source being measured, the C bias must be great enough that input peak voltages do not make the grid go positive at any time. In practice the bias is such that the normal d.c. plate current is very near zero; this near-zero current is prevented from going through the microammeter by a FIG. 3 Some calibration curves. Each curve was made with a different a. c. voltage input. Note how rapidly the plate current increases with small increases of plate voltage RAI I )I0 BI JVBOP ,OADC ATOR AST Y J / /, u / * H 1 1 1 1 f 4 / 1 1 1 1 E 199 2F=3 8 a.c.- ■-2 / a.c.=i t / f t I 1 *a t f f c.=o • / / / 1 — ✓ t / / / / 200 160 140 120 E 100 60 40 20 12 14 16 18 20 22 24 26 Ed 28 30 32 36 38 40 "bucking" voltage secured from a battery or a potentiometer across the A battery, as shown in Fig. 2. The high resistance prevents shunting the meter with the potentiometer. The reason for preventing the normal d.c. plate current going through the meter is so that the latter's range will not be limited by having to register both the normal current and the current produced in the process of measurement. Any type of tube may be used. For convenience, the small 199 type tubes are used in the Laboratory, and should be used with the voltages specified here. One simple form of vacuum-tube voltmeter used in the Laboratory is shown in the photograph on page 221 and another type, housed in a Ware radio cabinet, on page 223. The microammeter in the first picture (that at the top of page 221) is a Westinghouse instrument Model PX and has a maximum range of 200 microamperes. It is also possible to obtain this model with a full-scale deflection of 500 microamperes. These are particularly good instruments for this work, since the scale length of four inches makes it easy to read accurately, and the little button at the left, when pushed, removes a shunt which protects the meter from overload. In other words, the meter is always protected until one pushes the button, and in a vacuum-tube voltmeter of the type described here, this is a very valuable feature. The two Westinghouse meters list at about $35. Weston makes a model 301 meter in two ranges which are suitable for tube voltmeters, one with a full-scale deflection at 200 microamperes listing at $33 and another, a i-milliampere (1000 microamperes) meter at $12. Jewell makes similar instruments at about the same prices. In the photograph on page 223 a Weston 1 .5-milliampere (1500 microamperes) meter is shown. CONSTRUCTION OF THE VOLTMETER HPHE photograph on page 221 shows how sim* pie the vacuum-tube voltmeter may be. The small single-pole double-throw switch at the left is used to short the input by connecting the grid directly to the negative post of the C battery when circuits to be measured are being set up. There is no need to place the apparatus in such a small space as shown in this photograph, although the grid lead between voltmeter and external apparatus under test must be as short as possible, and well protected from other leads carrying a.c. voltages. A list of the apparatus used in the voltmeter shown in Fig. 2 follows. Any of these parts may be substituted by others that are well made, and none of the values of resistance, etc., are critical. One Tube Socket. Ri — 30-Ohm Rheostat. R2 — 400-Ohm Potentiometer. R3 — 20,000-Ohm Resistance. Ci — i-Mfd. Bypass Condenser. One S.P.S.T. Switch. 7 — Clips (or Binding Posts). One Binding Post, Insulated from the Baseboard. Microammeter. The experimenter who builds such an a.c. voltmeter for the first time must remember that he has in the circuit a very sensitive and, therefore, expensive meter, namely, the plate current reading device If the grid circuit of the tube is left open, or if any one of several accidents happen, the meter will be blown up, and all experiments will terminate in an abrupt and disheartening manner. Every step in the construction, calibration, adjustment, and operation must be watched with great caution, and adjustments should be