ENCYCLOPEDIA OF RADIO ELECTRONICS AND ELECTRICAL ENGINEERING Elimination of the effect of transistor sounding of powerful UMZCH. Encyclopedia of radio electronics and electrical engineering Encyclopedia of radio electronics and electrical engineering / Transistor power amplifiers Avid music lovers, musicians and sound engineers have long noticed that there is a difference in the sound of powerful tube and transistor AF amplifiers. With regard to the measured values of their parameters, transistor amplifiers are not inferior, and sometimes even superior to tube ones. But when listening to transistor ultrasonic frequencies, the so-called "transistor sound" often appears. It manifests itself in the distortion of the natural timbre of musical instruments and can be succinctly characterized as a loss of natural "lightness" of sound, insufficient "transparency" of sound, as well as a specific reproduction of the high-frequency components of the signal, expressed in the feeling of their "difficult" passage through the sound reproducing path. The conducted studies have shown that this effect in different amplifiers of the same class does not manifest itself at all in the same way. The researchers classified the amplifiers, ranking them in order of deteriorating sound and amplifying "transistor sound". As a result, Russian experts stated that the appearance of "transistor sound" is associated with the coefficient of non-linear distortion, with the proviso that all other parameters of the amplifiers are the same. This conclusion is confirmed by the results obtained by a number of Western researchers [1-3], showing a strong influence of the nonlinearity of the amplitude characteristic, estimated from the coefficient of nonlinear distortion of the signal. It should be noted that the quality of sound reproduction is negatively affected not only by non-linear distortions. To a much greater extent, this is due to the combination components of the signal spectrum, which arise due to the nonlinearity of the amplitude characteristic with simultaneous amplification of signals with different frequency spectrum [4]. When studying combinational components, the MEK recommendations were used to measure the so-called "TIM-distortions" (Transient Inlermodulation Distortion). Signals with frequencies of 3,18 kHz and 15 kHz were fed to the input of the amplifiers with the same amplitude, providing an output power with a level less than the nominal level by 3 dB. The test results confirmed the theoretical assumptions that the output signal of transistor amplifiers is richer in harmonics (the presence of about 11 harmonics is observed) than that of tube amplifiers (the spectrum has up to 5 harmonics), which affects the subjective perception of the sound picture. In addition, it turned out that the spectrum of combination frequencies of transistor amplifiers is "dense" than that of tube amplifiers. These features in the spectrum of harmonics and combinational components are, according to the authors, one of the main reasons for the appearance of "transistor sounding". From the foregoing, the obvious conclusion is that that the norms for the coefficient of non-linear distortion (Kni) of tube amplifiers are not applicable for transistor UMZCH. For them, admissible books should be much less. The same is true for the intermodulation distortion factor. Anticipating the difficulties of purposefully influencing the width of the spectrum of the harmonic components of the useful signal, the only way to deal with the "transistor sound" is to reduce the Knee to a value at which the influence of the combinational frequencies of the signal is not subjectively felt. This requires a method for assessing non-linear distortions, which makes it possible to unambiguously determine the threshold below which "transistor sound" does not appear. The method for assessing the quality of amplifiers using TIM distortion does not differ significantly from the well-known spectral method, but is not applicable in practice, since new specialized measuring equipment is required. As the studies presented in [6] show, the single-signal method is quite applicable for evaluating the Knee in any sound engineering system with a uniform frequency response, which is easily achieved in high-quality Hi-Fi amplifiers. The results below are obtained as a result of experiments carried out according to the single-signal method. Due to the inherent nonlinearity of transistors, the construction of amplifiers without the introduction of a special device to reduce nonlinear distortion is impossible. The most effective way to reduce KNI is the introduction of negative feedback (NFB). To avoid a number of troubles that every designer faces when developing an output stage with OOS [6, 7]. the following rules must be observed:
Best meets the specified requirements UMZCH, designed and built by Lynch Marshall [8]. This amplifier is comparable to tube amplifiers. The corresponding results are shown in the table. During testing, the amplifiers were switched on according to the scheme shown in Fig.1. Here U1 is a studio tape recorder. Z1 - multi-band equalizer. A1 and A2 are amplifiers whose sound quality is compared. In order not to disturb the purity of the experiment, there were no frequency filters in the loudspeakers that introduce phase distortions. Acoustic systems (of our own design) had loudspeakers produced by Gudmans, characterized by low non-linear distortion in the frequency range of 0,03 ... 16.5 kHz. As a signal source, we used a program recorded on studio equipment on an A4615-6P tape at a speed of 38,1 cm / s from a high-quality gramophone record played by an Otophon player built into the XL-1550 gramophone chassis of the Pioneer device. To avoid overloading the inputs of the amplifiers, signal levels were set so that even at peak output power it remained 3 dB below the maximum. When listening, the superiority of amplifier No. 1 over amplifiers No. 2 and No. 3 was felt in relation to the "purity" and "transparency" of the sound picture when transmitting the higher components of the sound spectrum. In addition, to achieve approximately the same balanced (timbre) sound, the EQ response of amplifier #1 was uniform, while when working with amplifier #2, a boost of + 10 dB was required in the frequency range from 1 to 16 kHz. Amplifier No. 3 was inferior in sound quality to all the others. Tube amps #4 and #5 have not been unanimously agreed upon, but they have been found to have no advantage over #1 amp. In this regard, additional tests of amplifier No. 1 were carried out when it was included in a two-way tube sound-reproducing complex with electromechanical feedback (EMOS) and with a bandwidth (according to sound pressure) of 0,016 ... 25 kHz. The block diagram of the installation is shown in fig. 2. As a test load of amplifier No. 1 (A2 in Fig. 2), a resistor divider R1-R2 was selected so as to obtain a transfer coefficient equal to 1. The test showed that the inclusion of amplifier No. 1 in an audio complex does not lead to the appearance of any "transistor tones " when playing a variety of music programs. It was found that the characteristics of UMZCH No. 1 almost coincide with the characteristics of UMZCH No. 2, but they have a significantly lower Knin - not exceeding 0.04% in the band 0,02 ... 20 kHz. This is the meaning of Kni. obviously, it is the desired boundary at which the "transistor sound *" disappears. Taking as a basis the principles for designing high-quality AF equipment, as well as a relatively cheap element base, the authors developed a power amplifier, the circuit of which is shown in Fig. 3. The pre-amplifier consists of an emitter follower on a VT1 transistor and a symmetrical push-pull cascade on VT2, VT3, covered by a local OOS due to emitter resistors R11 and R12 and a common OOS, wound up from the collectors VT2, VT3 through a divider R1-R2-RP3 to the base VT1. The OOS signal is added there with the input signal. Resistors R2 and RP3 simultaneously serve as an input signal divider. The gain of the pre-amplifier without OOS is about 100, Knee at the maximum input signal is about 0,15%. The introduction of OOS reduces the gain to about 5.5, and Knee - to 0.01%. Balancing of the cascade is carried out by the resistor RP8. The "buildup" cascade is assembled on transistors VT4, VT5 and VT6 according to a scheme similar to the preamplifier. The gain of this cascade without OOS is about 100, and Kni = 0,1 ... 0,15%. This is achieved through the use of transistors BD140/BD139 (without any selection of transistors by parameters). The VT4 emitter follower serves to increase the efficiency of the parallel OOS introduced from the amplifier output through the R14-R15-R20 divider. The cutoff frequency of the cascade is determined by the capacitances of the collector junctions VT5, VT6 and the value of C13. For the capacitance C13 indicated in the diagram, the cutoff frequency is approximately 35 kHz. The R16-C8 chain corrects the frequency response. The output stage according to the scheme is similar to the Brig 001-stereo amplifier. To avoid an increase in Knee and the appearance of "transistor sound", local feedback is used, implemented on resistor dividers R38-R39 and R40-R41-R42-RP44 with low resistances. As in the previous stage, the selection of transistors was not carried out. With the help of RP44 the output signal kn is minimized. Without OOS Knee in the entire band of audio frequency is 0.5 ... 0,7%, the gain is 2.7. The quiescent current of the output transistors is set to about 100 mA using the RP30. and setting "0" at the output is made by the resistor RP24. With a common negative feedback covering the "swinging" and output stages. Knee at maximum output power over the entire frequency range is 0,02% (measured by the compensation method). With the exception of the low-pass filter formed by the R14-C6 chain, in the "small signal" mode (a signal with a level of 0,1 of the nominal was applied to the input), the upper cut-off frequency of the amplifier was 1.8 MHz! To prevent self-excitation of the amplifier, a Bouchereau compensator - R54-L1 is installed at the output. Coil L1 (inductance - about 0,3 μH) is wound on R54 (along the entire length) with a wire of 0.8 (1,0) mm. The following replacement elements are possible in the amplifier: VT1, VT3, VT4, VT7, VT8 - VS546V, 2T3167V (C), VS107. KT315V(G); VT2, VT9 - VS556V, VS177V(S), 2T3307V(S), KT361B(G); VT5 - 2T9140C, KT814B; VT6 - 2T9139C. KT815V: VT10 -2T7638V. KT626V; VT11 -2T7637V, KT807B; VT12, VT13 - KD3442. 2N3442, 2N6259A, KD502. The VT7 transistor is smeared with heat-conducting paste and fixed on the radiator near VT12 or VT13 (at the top of the radiator). In conclusion, we can say that:
Literature
Authors: D.Kostov, V.Todorov See other articles Section Transistor power amplifiers. Read and write useful comments on this article. Latest news of science and technology, new electronics: The world's tallest astronomical observatory opened
04.05.2024 Controlling objects using air currents
04.05.2024 Purebred dogs get sick no more often than purebred dogs
03.05.2024
Other interesting news: ▪ New RECOM pulse converters with increased output current ▪ Electric car that absorbs carbon dioxide ▪ The Metaverse may be worse than social networks ▪ Is it easy for a whale to feed ▪ Women experience less pain when a man is around News feed of science and technology, new electronics
Interesting materials of the Free Technical Library: ▪ site section Electrical work. Article selection ▪ Tree of life article. Popular expression ▪ article A copy of which country with all cities and terrain exists in Minecraft? Detailed answer ▪ article Basic requirements for emergency preparedness ▪ article Egg in a bottle. Focus Secret
Leave your comment on this article: All languages of this page Home page | Library | Articles | Website map | Site Reviews www.diagram.com.ua |