The Vacuum Tube as an Amplifier
By B.F. McNamee
Radio for July, 1922
The type of vacuum tube used as an amplifier is the "hard" or high vacuum tube. The vacuum is so nearly perfect that there is no noticeable effect of gas molecules; in fact for practical purposes it may be considered a perfect vacuum. The "soft' tube, that is one containing a certain small pressure of gas and intended for use as a detector, can be used to a certain extent for amplification purposes. However this almost always results in distortion which would prohibit its use as an amplifier in radio telephone work, and its use is also restriced because it will not work with high plate voltage, thus limiting the strength of signals that can be obtained.
Fig. 1 and Fig. 2
The hard tube therefore is the only one that will be considered in this article.
In the soft tube we have both electrons and gas molecules acting as carriers of electricity, but in the hard tube the electrons are the only carriers and the action is thus simplified. When the filament becomes incandescent it throws off negative electrons. The B battery is connectd with is positive end toward the plate and this positive charge on the plate attracts the negative electrons. There is, therefore, a flow of electrons across the tube from filament to plate and returning to the filament through the meter and B battery shown in Fig. 1. As long as no charge is placed on the grid this current will remain at a steady value of about one or two milliampers with ordinary B battery voltages.
The grid is a sort of screen interposed between the filament and the plate. The electrons in going from the filament to the plate pass through the openings in this screen. When electric charges are placed on the grid it acts like the faucet in a water pipe since it controls the flow of current rhrough the tube. When a negative charge is placed on the grid, it repels the electrons coming from the bulb and a smaller number of them will reach the plate. If this negative charge is sufficiently great all the electrons from the filament will be repelled and driven back to the filament, and the meter which shows the plate current will read zero. If, on the other hand, a positive charge is placed on the grid, it will accelerate the electrons, and larger number will reach the plate. This will increase the current through the meter.
The grid is charged positively or negatively by connecting the battery between the points X and Y in Fig. 1. If points X and Y are connected by a wire, and Y is connected to the negative end of the filament, the grid is said to have zero charge. By changing the voltage of the battery connected between X and Y and marking down the corresponding values of current shown by the meter we obtain the curve of Fig. 2. We note that when X and Y are connected by a wire, that is, when there is zero charge in the grid, there is a current of 0.8 of a milliampere in the plate circuit. This value of current holds good for only one particular adjustment of filament and B battery on one particular tube; we might make this current almost anything we pleased.
Note that as the voltage is increased in the negative direction the current in the plate circuit falls very suddenly until we have reached a negative charge on the grid of 4 volts, corresponding to point M on the curve. As the voltage is increased in the negative direction beyond this point the plate current falls very gradually to zero at about 10 volts.
On the other hand, if the plate voltage is increased in the positive direction the plate current rises suddenly at first until about 4 volts positive has been reached, corresponding to the plate current of 1,5 milliampers. If we increase the positive charge of the grid beyond this point there will be hardly any corresponding increase in plate current. This is because all the electrons given out by the filament are now reaching the plate and consequently nu further increase in plate current can take place except the filament is made hotter. This is known as the saturation part of the curve.
Suppose that a small alternating voltage is connected between points X and Y in Fig. 1. If the voltage on the grid is made to fluctuate from 2 volts positive to 2 volts negative it will be seen thet the needle of the milliammeter will show that the plate current is rising and fallinf between tha values 1.2 to .6 milliamperes.
The usual method of supplying an alternatingvoltage to the grid of a vacuum tube is by means of a transformer, as shown in Fig. 3. The primary of the transformer may be connected to a telephone line and the alternating current in it will in that case consist of voice currents. The transformer is usually of the steo-up type in order to supply as great a change of voltage on the grid as is practically possible. Since the grid rewuires little or no current we can have large voltages from a source supplying very little power. This is the case in long telephone lines where nearly all the power is expended in overcoming resistance losses in the line.
Fig. 3 and Fig. 4
With such a voice-controlled voltage applied to the grid, the current in the plate circuit of the tube will fluctuate accordingly, and if a pair of telephone receivers are used in place of the milliammeter in Fig. 1, the speech of the telephone line will be clearly heard. The result will be much stronger than if the telephone receivers are connected directly in the telephone line, because the energy now used to produce the sound waves comes from the B battery and is simply controlled by the minute amount of energy in the telephone line. The grid circuit of the vacuum tube is for this reason called the control circuit or input circuit, and the plate circuit is known as the output circuit. A very small amount of energy in the control circuit will release in the output circuit a comparatively large anount from the B battery.
In telephone work another transformer would be connected in place of the telephones and this would supply the energy to the continuation of the telephone line. Of course this would provide for one way communication only, but rather complicated hook-up has been devised, generally using two tubes which provides for amplification in both directions and is used in all long distance telephoning today.
Fig. 4 shows the curve obtained from some types of amplifier tube with certain adjustments of filament and B battery. If this tube were used with the connections in Fig. 3, the amplified sounds would be distorted, because a positive charge of let us two volts would increase the plate current but very little, while a negative charge of the same amount would cause a comparatively large decrease. In order to amplify without distortion it would be necessary that the increase in this case should equal the decrease. This could be brought about by connecting a battery of about three volts between the filament and the secondary of the transformer with its negative end toward the latter. This would bring the plate current to point L on the curve of Fig. 4. If the grid voltage now fluctuates two or three volts in either direction, the corresponding increase and decrease in plate current will be equal and there will be no distortion. Such a battery connected in the grid circuit is called a "C" battery or "bias" battery. Since it supplies little or no current it may consist of the smallest size of flashlight calls.
If more amplification is required than that given by one tube, the telephones of Fig. 3 may be taken out and the primary of another transformer connested in their place. The secondary of tis transformer is connected to the grid and filament of another tube, which will amplify the results of the first tube. This action may be continued through four or five tubes in succession, and is known as cascade amplification. Very careful design is necessary in such multi-stage amplifiers in order to prevent distortion.
The matter of "howling" in amplifier circuits can be entirely avoided. This subject will be taken up in the next article of this series, which will deal with the vacuum tube as an oscillator.