Development of Transistor Power Amplifiers and Intermodulation Distortion

                     Development of Transistor Power Amplifiers and Intermodulation Distortion

With the gradual maturity of semiconductor technology, the varieties of high-current and high-voltage transistors are increasing day by day, and more and more power amplifiers use OCL circuits or OTL circuits without output transformers. The initial high-power PNP tube is a germanium tube, while the NPN tube is a silicon tube. The characteristics of the two are very different, and the symmetry of the circuit is very poor. People mostly use the quasi-complementary circuit shown in Figure 2. Through low-power The silicon tube Q1 is compounded with a high-power NPN silicon tube Q2 to obtain a high-power tube with a polarity similar to that of the PNP tube, which reduces the distortion caused by poor symmetry of the circuit. At the end of the 1960s, when high-power PNP silicon tubes were commercialized, complementary symmetrical circuits were widely used. The progress of components has made a qualitative leap in the technical indicators of transistor power amplifiers. In terms of subjective sound quality evaluation, it has also changed people's views on transistor power amplifiers in the past. Whether it is in hall sound amplification, radio program production or home playback, transistor power amplifiers Power amplifiers have been widely used, and for the first time surpassed tube power amplifiers by an overwhelming advantage in quantity. Among the commercialized transistor amplifiers, some eye-catching famous ones have appeared one after another, such as JBL's SA600, Marantz complementary symmetrical circuit MOdel15 and so on.

Although there are still a large number of supporters of electronic tubes, people can look at transistor amplifiers fairly after all. They think that transistor amplifiers have a wide frequency response and delicate layers. Compared with electronic tube amplifiers, they have a unique strength. question.

The proposal of transient intermodulation distortion is a leap in understanding. In the 1970s, one of the most striking things occurred in the development history of power amplifiers. This is the proposal of transient intermodulation distortion (Transientlntermodulation) and its measurement method. . In 1963, when an engineer at the Helvar factory in Finland was making a transistor amplifier, due to wiring errors, the amount of negative feedback in the circuit was reduced. Later, he unexpectedly found that the sound quality after the amount of negative feedback was reduced was very good, and the objective technical indicators were relatively good. Poor, and the line after the error is corrected, although the technical indicators have improved, the sound quality is significantly lower than that of the wrong connection. This phenomenon attracted the attention of Mr. Otala of the same factory at that time. Afterwards, he conducted careful research on it, and in 1970 he first published a paper on Transient Intermodulation Distortion (TIM) of transistor power amplifiers. By 1971, Dr. Otala and his research group had published more than 20 papers on TIM distortion theory, which aroused widespread repercussions from people in the field of electroacoustic quasi-complementary circuits.

The general idea of transient intermodulation distortion is this:

In a directly coupled transistor amplifier circuit, in order to obtain a small harmonic distortion and a wide and flat frequency response, negative feedback as deep as 40dB to 60dB is usually applied to the overall circuit. If the open-loop distortion of the amplifier before adding negative feedback If it is 10%, then after adding 40dB of negative feedback, the distortion can be reduced to 0.1%, which is difficult to achieve with the efficiency of the tube. The transistor power amplifier has to apply 40dB. 60dB negative feedback, so for an amplifier with a gain requirement of 26dB, its open-loop gain must reach 66, 86dB.

The introduction of deep negative feedback under such a high gain will inevitably cause self-excited oscillation in the circuit, so phase compensation is required. Generally, a small capacitor C is connected between the collector and base of the push transistor to destroy the self-excited oscillation. phase conditions, forming the so-called "lag compensation".

When a transitional pulse with a very short duration is input to the input terminal of the amplifier, since the capacitor C needs charging time, the collector voltage of the push tube can reach the maximum value after a period of delay, as shown in Figure 4. Obviously, during the charging and discharging of the capacitor C, the output voltage V is. It will not reach the proper voltage value, and the input stage cannot get the proper feedback voltage Vf. Therefore, the input stage will be negative when the transition pulse passes through the input stage. Feedback is out of control, resulting in serious overload of the input stage, and the output will be severely clipped, causing transient pulse distortion. If there is a sinusoidal signal superimposed on the transition pulse waveform, the output will also get a lot of intermodulation frequency components that do not exist in the input signal spectrum, which is TIM distortion.

TIM distortion is also closely related to music signals, and program signals with high volume and high frequency can easily induce TIM distortion. Severe TIM distortion is reflected in the sense of hearing as high-frequency cross-selection distortion, while weaker TIM distortion gives people an unpleasant feeling of "metal sound", resulting in deterioration of sound quality. So far, there are still disputes about TIM distortion in the audio industry, but this is after all the deepening of people's understanding. It has caused a fundamental change in the design thinking of amplifiers, that is, paying more attention to the dynamic performance of amplifiers rather than just being satisfied with static technical indicators. improvement.