Gain Control

Author: N.H. Crowhurst

It is easy enough by one means or another to get as much amplification as we want. In fact, if we are not careful, we will go on adding amplification and end up with too much. For this reason, we need some means of controlling amplification, called a gain control or a volume control. A simple volume control consists of a resistor with a slider riding on it. It is called a potentiometer, because it is a device that allows the potential at the slider to be varied.

The first idea for a volume control was to put this potentiometer across the output of the amplifier and take the connection to the loudspeaker from the slider and one end. This varies the amount of power delivered to the loudspeaker, but there is one serious disadvantage. If the amplification is too great, the amplifier will distort the signal, and turning down the control will merely adjust the loudness of the distorted program.

The next obvious place to put the volume control would be at the input end of the amplifier. Turning down the volume would then eliminate distortion. However, all amplifiers have a limited dynamic range. At the input end, the problem is noise, not distortion. Putting the gain control at the front end of the amplifier, means the loudspeaker gets all the noise amplified up from the input stage (which is usually the point at which noise limits dynamic range). For this reason, the best place to connect a gain control is somewhere in the middle of the amplifier.

Volume control in middle cuts down tube hiss from the input stage and distortion from the output stage

How a variable-mu tube works

Gain can also be adjusted by altering the operating conditions of a tube, especially if we use a special kind of tube with a considerable amount of curvature in its characteristic. These tubes do not have uniform spacing of the wires that make up the control-grid. This construction produces a tube whose transconductance varies considerably with grid bias voltages. When the grid voltage is only slightly negative, it has practically no effect on electrons passing through the wider part of the grid mesh, but it does influence the number of electrons passing through the closer part of the grid mesh. (In this range, the tube has a high transconductance.)

Making the grid more negative prevents any electrons at all from passing through the closer part of the grid mesh. Those passing through the wider part are subject to less control, and the tube acts as if it had a much lower transconductance. If the change in the spacing of the grid wires is gradual, adjustment of the grid voltage will give a very smooth change of gain as the bias is changed.