Transient Distortion

Author: N.H. Crowhurst

Testing amplifier response using square wave

This is still another kind of distortion which can be subdivided into further groups. It is not indicated in the majority of amplifier specifications. An amplifier may have a perfectly flat frequency response throughout the audio range, which should indicate that it will give a faithful reproduction of the input waveform at the output. It may show quite low harmonic distortion and yet when a square wave is amplified, the wave may become considerably distorted by a ringing at the corners of the square.

A square wave can be considered as the synthesis of a fundamental with a whole range of odd hamonics

A square wave can be considered as a synthesis or combination of fundamental with a whole range of odd-numbered harmonics. If all these are amplified uniformly, surely the output waveform should still be square? This is true, but it does not take into account possible effects due to time delay in the amplifier, which may not be uniform at all frequencies. Every bit of stray capacitance from a plate or other electrode to ground at different points in the amplifier causes a slight time delay in the amplified audio. This adds up on the way through the amplifier.

If the time delay to all components of the audio waveform is the same, the output waveform will still be a square wave, but if it is different for the higher frequency components than it is for the low-frequency fundamental and its lower harmonics, the waveform gets altered. This is one way in which ringing occurs.

Another kind of transient distortion occurs when the amplitude of the audio suddenly changes. Suppose that a sine wave is amplified, but is stepped up and down at intervals. This will cause the output waveform to step up and down at intervals, which will cause the output voltage to step up and down. If the amplitude of the sine wave is stepped up and down in such a way that the outline (or envelope) follows a square waveform, then the output should faithfully reproduce this.

Marty amplifiers are not satisfactory in this regard. When a larger audio voltage is being amplified, the output tubes draw more current, which may alter the bias condition. This means that the supply voltages at different points in the circuit will change. The change will take place according to the time constants of the resistances and capacitances in the supply unit, which may not be the same for all the changing voltages. Consequently the gain of the amplifier may go up and down again or down and up again after a sudden change in the amplitude of the audio. This results in an envelope at the output that is different from the envelope at the input. Unfortunately these effects can prove quite severe, even with an amplifier whose specification, using the other methods of test, tells of quite good performance - extremely low distortion and very good frequency response.

Thus amplification is far from being the simple matter we started out by supposing. There are many ways in which an amplifier can distort a composite audio program. Whichever method of specifying these is used, it becomes quite an involved matter to give a statement that is completely satisfactory for comparison purposes.

Transient distortion