Audio amplifier distortion and common improvement methods

Whether it is electrical distortion or acoustic distortion, according to the nature of the distortion, there are mainly two kinds of frequency distortion and nonlinear distortion. Among them, the relationship between the amplitude and phase of each frequency component of the signal changes, and only waveform distortion occurs, and no new frequency component is added, which belongs to linear distortion. Harmonic distortion (THD), intermodulation distortion (IMD), etc. can produce new frequency components or modulation products of each frequency component. These extraneous products are extremely dissonant with the original signal, causing sound distortion and rough harshness. These distortions are Non-linear distortion.

Here, we discuss harmonic distortion, intermodulation distortion, transient intermodulation distortion (TIM), AC interface distortion (IHM), etc., respectively.

1 Harmonic distortion

Harmonic distortion is a type of distortion caused by nonlinear components in the power amplifier. This distortion causes the audio signal to generate many new harmonic components, which are superimposed on the original signal to form a waveform-distorted signal. The sum of the distortions caused by each harmonic is the total harmonic distortion, and its value is usually expressed as a percentage of the ratio of the root mean square value of all harmonics in the output signal to the effective value of the fundamental voltage.

Here, the fundamental signal is the input signal, and all harmonic signals are the harmonic signals introduced by nonlinear distortion. Obviously, the smaller the percentage, the smaller the harmonic distortion and the better the circuit performance. At present, the harmonic distortion of Hi-Fi amplifiers is generally controlled below 0.05%. The harmonic distortion of many high-quality power amplifiers is already less than 0.01%, while the harmonic distortion of professional audio amplifiers is generally controlled below 0.03%. In fact, when the total harmonic distortion is less than 0.1%, it is difficult for the human ear to distinguish.

It should also be noted that for a specified audio power amplifier, for example, the total harmonic distortion index of an audio power amplifier is expressed as THD<0.009% (1W). At first glance, it seems that the total harmonic distortion is very small, but it is only the total harmonic distortion when the output power is 1W, which is different from the total harmonic distortion value obtained under the measurement conditions required by the relevant standards. Therefore, when marking the total harmonic distortion index of the audio power amplifier, the measurement conditions are generally indicated.

As we all know, the human hearing system is extremely complex, and sometimes power amplifiers with small harmonic distortion are not as good as those with high harmonic distortion. The reasons for this phenomenon are various. Among them, it is directly related to the degree of influence of each harmonic component on the sound quality. Although the steady state test data of the stone machine and the amplifier are the same, people always feel that the bass of the amplifier is mellow, bright and round, and the treble is slender and clear. But high-frequency hair, hair stiff, cold tone.

Through spectrum analysis, it is found that the stone machine contains a large number of odd harmonics. The odd harmonics cause a harsh and unpleasant sensation to the human ear; the bile machine contains rich even harmonics, and the human ear is not sensitive to even harmonics. In addition, the human ear has low resolution for even-order harmonic distortion, but is very sensitive to higher-order harmonics, which is also one of the important reasons for the above phenomenon.

The main ways to reduce harmonic distortion are:

1) Apply an appropriate amount of voltage negative feedback or current negative feedback; 2) Use high fT, small NF and good linear amplification components; 3) Improve the consistency of the tubes in each unit circuit as much as possible; 4) Use Class A Amplify the way, choose excellent circuit program; 5) Improve the power reserve of the power supply, improve the filtering performance of the power supply.

2 Intermodulation distortion

Two or more signals of different frequencies are intermodulated after the amplifier or when the speaker sounds to produce sum frequency and difference frequency, and each harmonic combination produces sum frequency and difference frequency signal. These newly added frequency components constitute a non- Linear distortion is called intermodulation distortion. Generally, two high and low frequency signals with a certain amplitude (take 4:1 more) are mixed into the circuit, and the ratio of the root mean square value of the newly generated nonlinear signal to the amplitude of the original higher frequency signal is measured Intermodulation distortion, that is, the magnitude of intermodulation distortion, can be expressed as a percentage of the intermodulation product level and the rated signal level. The larger this value, the greater the intermodulation distortion.

Obviously, the magnitude of intermodulation distortion is related to the output power. Because these newly generated frequency components have no similarity to the original signal, the small intermodulation distortion is also easily noticeable by the human ear, and it sounds sharp and harsh, and is accompanied by the phenomenon of "acoustic staining". In other words, the impact of intermodulation distortion will make the sound field of the entire playback system lack a sense of hierarchy and decrease the clarity. In Hi-Fi amplifiers, I always hope that the lower the intermodulation distortion, the better. It is very difficult to achieve this, so high-fidelity amplifiers require this value to be less than 0.1%. Of course, the comparison between the stone machine and the amplifier has a larger intermodulation distortion, which is one reason why the sound of the stone machine is not as sweet as the amplifier.

The methods to reduce intermodulation distortion are:

1) Adopt electronic frequency division to limit the working bandwidth of the amplifier circuit or speaker;

2) Add a high-pass filter to the input end of the audio power amplifier to eliminate sub-low frequency signals;

3) Choose a tube or circuit structure with good linearity.

3 Transient distortion Transient distortion is an important indicator of modern acoustics, which reflects the ability of the power amplifier circuit to keep track of transient jump signals, so it is also called transient reflection. In a high-fidelity system with transient distortion, the output music signal lacks layering and transparency. Generally, the causes of transient distortion are:

1) The role of the reactive components in the circuit is too large, and the frequency range is not wide enough; 2) The action of the speaker vibration system cannot keep up with the changes of the transient electrical signal.

There are two main forms of transient distortion, namely transient intermodulation distortion and distortion caused by low conversion rate (SR).

A. Transient intermodulation distortion

When a pulsed transient signal is input, due to the presence of capacitance in the circuit (such as hysteresis compensation capacitance, tube-to-pole capacitance, etc.), the output terminal cannot immediately obtain the proper output voltage (ie, phase lag) and the input stage cannot be obtained in time. Due to the negative feedback, the amplifier is in an open-loop state at this instant, which overloads the input stage instantaneously. The input voltage at this time is dozens of times higher than normal, resulting in severe clipping of the input stage. Wave distortion is called transient intermodulation distortion. It is essentially a transient overload phenomenon.

Due to the strong anti-overload ability of the amplifier, the low magnification, no deep-level negative feedback, and only some local negative feedback, it is not easy to produce transient intermodulation distortion. The general stone machine adopts a large loop deep negative feedback network to meet the requirements of low distortion and wide frequency band. It can be seen that the transient intermodulation distortion mainly occurs in the stone machine. In addition, programs with high volume, high frequency, and large dynamic range are most likely to produce transient intermodulation distortion. The reason is that the maximum time change rate of music around the zero signal level will make the sound not completely clear, especially the low-end stone machines often appear in the high-frequency part, producing a sharp, harsh feeling, so-called "Transistor sound" and "metal sound".

Transient intermodulation distortion is a dynamic index proposed in the 1970s, mainly caused by deep negative feedback inside the audio power amplifier. It is recognized as the culprit that affects the sound quality of stone machines, leading to "transistor sound" and "metal sound", and people attach great importance to this.

Improving TIM can start with its formation mechanism, and the commonly used methods are:

1) Control the open-loop gain and negative feedback of the amplifier to about 50dB and 20dB, respectively;

2) Use high fT tubes, use fT tubes greater than 100MHz in the front stage, and fT tubes at the final stage power should be greater than 20MHz, try to widen the open-loop frequency response of the circuit, increase the current negative feedback of each level, and cancel the large loop Road negative feedback. At present, the final stage expansion circuit of some power amplifiers (such as the bell JA-100) does not involve loop negative feedback, one of its purposes is this;

3) Adopt a fully complementary symmetric circuit to increase the working current of the power output stage, and add a buffer amplifier stage before the output stage to improve the transient response of the circuit;

4) Cancel the phase lag capacitor and change the lag compensation to lead compensation, that is, without a lag compensation capacitor, a small capacitor of appropriate capacity is connected in parallel with the large loop feedback resistor;

5) Appropriately increase the static current of the input stage, increase its dynamic range, and set a low-pass filter in its input circuit to eliminate high-frequency clutter signals above 80 kHz to prevent high-frequency interference signals from causing transient overload of the input stage.

B. Distortion caused by low conversion rate

Conversion rate refers to the ability of audio equipment to track or respond to burst sound signals or pulse signals, and is an important parameter that reflects the transient strain capability of the power amplifier circuit. The transient distortion caused by the low conversion rate is caused by the change in the output signal of the amplifier that cannot keep up with the rapid change of the input signal. If a sufficiently large pulse signal is input to the amplifier, the maximum rate of change of its voltage should be the ratio of the voltage rise value to the required time, the unit is how many volts rise per second, written as a digital expression SR=V/μs.

For high-fidelity amplifiers, SR directly affects the transient response and response speed of the amplifier. Amplifiers with high SR values have good resolution, layering, and positioning, and good listening, especially for replaying popular music. The value of SR is related to the output voltage of the power amplifier and the output high-frequency cutoff frequency. The SR value is larger if the output power is large; the SR value is also high when the high-frequency cutoff frequency is high. The SR value of high-quality power amplifiers can reach 100V/μs . In order to improve the SR value of the power amplifier, ultra-high-speed, low-noise tubes are usually used. However, if the SR value is too high, it is easy to make the circuit self-excited and its stability deteriorates.

In addition, the SR value of the pre-stage circuit should not be higher than that of the post-stage circuit, otherwise it will easily cause transient intermodulation distortion. By the way, the SR of the power amplifier can be estimated by an oscilloscope. The method is to feed the audio power amplifier with a square wave signal as the input signal, and the time required for the leading edge of the output signal waveform to rise to the rated value. The result is V/μs The representation is the size of the conversion rate. Obviously, if the audio power amplifier can process square wave signals well, it shows that it has a good conversion rate and a wide frequency characteristic.

4 AC interface distortion

The AC interface distortion is caused by the back EMF of the speaker being fed back to the circuit through the line. The methods to improve this distortion are: 1) reduce the number of circuit stages and appropriately increase the static working current of the circuit; 2) choose a suitable speaker to make the damping coefficient more reasonable; 3) adopt a large-capacity high-quality power transformer and increase it appropriately For the capacity of the filter capacitor, connect a small-capacity CBB capacitor in parallel with the filter capacitor.

In addition, due to improper selection of the DC operating point of the circuit or the low quality of components, some other nonlinear distortions, such as crossover distortion and clipping distortion, can occur, which can cause harmonic distortion and intermodulation distortion. Crossover distortion is also known as crossover distortion. It is for push-pull power amplifiers, which is mainly caused by the non-linearity of the initial conduction of the power tube in class B push-pull power amplifiers, especially in the case of small currents. The current produces nonlinear distortion at the junction, and the smaller the signal amplitude, the more severe the distortion.

Wave distortion is caused by the insufficient dynamic range of the power amplifier tube. The large signal is clipped due to saturation conduction. Clipping distortion generates a large amount of ultrasonic waves, which makes the sound blurry and jittery. A common method to reduce crossover distortion is to appropriately increase the DC operating point of the push-pull output tube; and measures to improve clipping distortion generally increase the linear operating range of the circuit appropriately.