## FIG

Let's look at this a little more closely. Take an ordinary amplifier and feed some ac into the grid circuit. What do you have at the plate? If you answered 'more ac/' go to the foot of the class. Your amplified ac is there, to be sure, but the dc plate voltage is also present— and if the amplifier is working within its limits, the absolute voltage at any instant will never go below zero so the voltage at the p.ate cannot actually "alternate."

Now couple off the amplified signal from the plate, through a capacitor. What do you have now? The dc cannot pass through the capacitor, and as a result only the ac component gets through. It swings equally above and below zero; it is true ac.

What does all this have to do with modulators? To answer this, we have to draw a couple of pictures of "transfer characteristics/' The "transfer characteristic/' for t iiose un  Lu« * feíni dilatar ör o^r «fl familiar with the term, is simply a graph which shows output voltage of a circuit in relation to input voltage. For a triode amplifier, it would read instantaneous plate voltage on the vertical scale in relation to instantaneous grid voltage on the horizontal scale (Fig, 1),

Taking that amplifier in Fig. 1, let's examine it. The graph says that when the grid voltage is minus 3, plate voltage is 100. When grid voltage is minus 2} plate voltage drops to 50, \Y;ien grid voltage is minus 4? plate voltage rises to 150.

Such an amplifier is linear, in that the output voltage changes the same amount for equal changes of input voltage. What goes into a linear amplifier comes out unchanged.

Now, let's put two ac tones into the input of this amplifier (Fig. 2). The two tones are completely separate, and as a result neither of them is true ac; the higher-frequency tone "rides" the lower-frequency one, So far as the higher tone is concerned, the lower tone is its zero-voltage reference. But since the amplifier is linear, the tones come out unchanged.

At this point, let s take a look at the transfer characteristic of a perfect diode. It's shown in Fig. 3, Note that, unlike our amplifier, this characteristic is not a straight line- It has a sharp break in it; the sharper the break, the better the diode.

Now, in Fig. 4, let's apply those same two ac tones to the diode, When the input signal swings positive, the diode conducts and an output signal appears. When it swings negative, nothing happens. Thus, the output signal is not a replica of the input. The diode is said t i i a i * »>

to be non-linear.

Note that in all four illustrations, the output signal is not true ac. All have some dc components present. However, passing them through a capacitor will remove the dc and the output will then be true ac.

The output of the diode* when treated in this manner, becomes completely true ac; it's