29.06.2020
CRT TV cascade low frequency amplifier circuit. Zero-axis unch with tnd cascade. Energy consumption of the power supply
1. Review of amplifiers.
Most amplifiers consist of several stages that provide sequential amplification, usually called stages. The number of installed cascades depends on the required values of the gain factors and on the unit (intrinsic) gain factors of the discrete elements that make up the cascade.
A cascaded amplifier circuit can be thought of as functionally distinct amplification stages: pre-amplification, intermediate amplification and output (power) amplifier.
The preamplifier provides direct connection between the signal source and the amplification device. Therefore, the most important requirement that it must meet is minimal attenuation of the input signal. To do this, the preamplifier must have a large input resistance, provided that this resistance must be substantially the same as the resistance of the signal source. In this case, changes in the input voltage of the amplifier will tend to change the emf. source in its input circuit. The main requirement for the preliminary stage (amplifier) is to ensure the greatest amplification of the input signal with minimal distortion. The preamplifier as a discrete element is also called the input stage.
The intermediate amplifier acts as a buffer stage between the pre-amplifier and output amplifier. Its main task is to match the output of the input stage with the input of the output amplifier (power).
The output stage is designed to obtain the output amplification device, power that ensures the operation of a load device that performs certain functions. Therefore, unlike the preliminary and intermediate cascades, output power which is relatively small, the main parameter of the output stage is efficiency.
Transistor power amplifiers used in practice are classified into single- and push-pull. Single-ended power amplifiers are used to operate load devices whose power is a few watts. At high power values of load devices, push-pull amplifiers are used.
It should be noted that the presence of three different types of functional cascades - preliminary, intermediate and output - are not mandatory. There are amplifiers in which there are no clearly defined demarcation marks for the preliminary and intermediate stages; they can be combined in one stage. The same applies to the intermediate and output stages, which can also be combined.
Circuits of amplifier stages can be made in a variety of options. They may differ in the number and mode of operation of the amplification elements used when amplifying the alternating signal. There are several fundamentally different operating modes of the amplifier, called amplification classes:
a) class A - the current in the output circuit of the amplifier (transistor) flows during the entire period of change in the input signal voltage; the resting point is in the middle part of the load characteristic; the mode is characterized by low efficiency (no more than 0.5) and a low value of the nonlinear distortion coefficient kf;
b) class B - the current in the output circuit of the transistor flows only during half the period of change in the voltage of the input signal, while the resting point is actually in the cutoff mode of the transistor; this class is preferred for use in medium and high power amplifiers; The efficiency of the cascade can reach 0.7 or more in this class, however, it has the highest nonlinear distortion coefficient of all classes, due to the step at the output of the cascade;
c) class AB - the current in the output circuit of the transistor flows for more than half the period of change in the input signal voltage; the rest point is below the midpoint of the load characteristic; the class has become widespread, since with high efficiency it provides small nonlinear distortions of the output signal;
d) class C - the current in the output circuit of the transistor flows over an interval less than half the period of change in the input signal voltage; common in powerful resonant amplifiers, but the parameters are close to class B;
e) class D – a mode in which the transistor of the cascade can only be in the on state (saturation mode) or off (cut-off mode); The efficiency of such an amplifier is close to unity; most common - in digital circuits and transistor switches.
The choice of one or another mode of operation of the amplifier stage is determined based on the required values of the nonlinear distortion factor kf and efficiency.
The main direction in the development of modern discrete amplification elements is the study of their main characteristics, such as amplification quality, efficiency, weight and size indicators, etc. In integrated designs, the most important indicators are the dimensions of the elements and their reliability. Typical dimensions of a logical transistor element in modern processors are 25-13 microns. Particular prospects in this direction are molecular and atomic nanoassembly, that is, the actual limit is in units of nanometers.
2. BLOCK DIAGRAM OF THE AMPLIFIER
Structural scheme amplifier is built on the basis general principles construction of a ULF (low frequency amplifier). In accordance with this, the amplifier has an input stage, several pre-amplifier stages and an output stage. To ensure thermal stabilization of the amplifier's rest mode and the required gain, the amplifier is covered by a negative feedback, while the type of OOS depends on the input stage circuit.
The block diagram of the amplifier is shown in Figure 1.
Figure 1. Block diagram of the amplifier.
where VxK is the input stage;
KPU – pre-amplification stage;
VK – output stage;
NFE – negative feedback.
The amplifier works as follows. The input signal is supplied to the input stage of the input stage and is amplified in voltage. From the output of the input stage, the signal goes to the input of the pre-amplification stages of the CPU. From the output of the last preliminary stage, a signal with a voltage amplitude close to Unmax is supplied to the input of the output stage of the VC, amplified by current and power, and transmitted to the load.
3. DEVELOPMENT OF THE PRINCIPLE DIAGRAM OF THE AMPLIFIER.
3.1Selection of operating mode and output stage circuit.
According to the technical specifications, the coefficient of nonlinear distortion should be no more than 0.12%, and the efficiency should not be lower than 45%. These conditions correspond to the operating mode of the output stage in class AB with the introduction of negative feedback.
Since the power that needs to be transferred to the load from the output stage is not large (power in the load is 50 W), the output stage, according to operating class AB, must be built using a push-pull circuit.
The schematic diagram of the output stage is shown in Figure 2.
The output stage is assembled using transistors VT6...VT11. Transistors VT6 and VT10, as well as VT7 and VT11 are assembled, respectively, according to the circuit composite transistor. This circuit solution is determined by the technical specifications, according to which Circuit efficiency must be at least 45%. Without the required transmission coefficient of the output transistors, this condition is not met. The corresponding calculations will be given below.
Pre-amplification stages General information. The preamplifier amplifies the voltage or current fluctuations of the signal source to the values that must be applied to the input of the final stage to obtain the specified power in the load. The preamplifier can be single- or multi-stage. Transistors in the pre-amplification stages are switched on with an OE, and the lamps are switched on with a common cathode, which allows for the highest gain. Turning on a transistor with OB is advisable when input stages, operating from a signal source with low internal resistance. To reduce nonlinear distortion in pre-amplifier stages, mode A is preferred.
- Based on the type of connection between the stages (with multi-stage amplifiers), amplifiers are distinguished with capacitive,
- transformer
- galvanic coupling (DC amplifiers).
Capacitively coupled amplifiers. Amplifiers with capacitive or CN-coupling are widely used. They are simple in design and setup, cheap, have stable characteristics, are reliable in operation, and are small in size and weight. Typical amplifier circuits using transistors and capacitively coupled tubes The frequency response of a capacitively coupled resistor stage can be divided into three frequency regions: lower low frequencies, mid midrange and upper high frequencies. In the low-frequency region, the gain Kn decreases (with decreasing frequency) mainly due to an increase in the resistance of the interstage coupling capacitor Cp1. The capacitance of this capacitor is chosen to be large enough, which will reduce the voltage drop across it. Typically, the low-frequency range is limited by the frequency fH, at which the gain is reduced to 0.7 of the mid-frequency value, i.e. Kn=0.7K0. In the mid-frequency region, which makes up the main part of the amplifier's operating range, the gain Kо is practically independent of frequency. In the high-frequency region fB, the decrease in gain Kb is due to the capacitance Co=/=Cout+Cm+Cwx (where Cwx is the capacitance of the amplifying element of the cascade; Cm is the installation capacitance, Cwx is the capacitance of the amplifying element of the next cascade). They always try to minimize this capacitance in order to limit the signal current through it and provide a high gain. Calculation of a resistor pre-amplifier stage. Initial data: amplified frequency band fn-fv = 100-4000 Hz, frequency distortion factor MH Transformer coupled amplifiers. Transformer-coupled pre-amplification stages provide better matching of amplifier stages compared to resistor-capacitively coupled stages and are used as inverses to supply a signal to a push-pull output stage. Often a transformer is used as an input device. Circuits of amplifier stages with serial and parallel connection of a transformer are shown in. The circuit with a series-connected transformer does not contain resistor RK in the collector circuit, therefore it has a higher output resistance of the cascade, equal to the output resistance of the transistor, and is used more often. In a circuit with a parallel-connected transformer, a transition capacitor C is required. The disadvantage of this circuit is the additional loss of signal power in the resistor RK and the reduction in output resistance due to the shunting action of this resistor. The load of the transformer stage is usually the relatively low input impedance of the subsequent stage. In this case, step-down transformers with a transformation ratio n2=*RB/R"H are used for interstage communication The frequency response of a transformer-coupled amplifier has a reduction in gain in the low and high frequencies. In the low-frequency region, the decline in the cascade gain is explained by a decrease in the inductive resistance of the transformer windings, as a result of which their shunting effect of the input and output circuits of the cascade increases and the gain K=Ko/ decreases. At medium frequencies the influence of reactive elements can be neglected. In the high-frequency region, the gain factor is affected by the capacitance of the collector junction C and the leakage inductance ls of the transformer windings. At a certain frequency, capacitance Sk and inductance Is can cause voltage resonance, as a result of which at this frequency a rise in the frequency response is possible. Sometimes this is used to correct the frequency response of an amplifier. Final stages of low-frequency amplifiers
Single ended amplifiers
Single-ended amplifiers in tube receivers are used with an output power of no more than 4...5 W. For high output powers, push-pull amplifiers are usually used. Fig.1
To ensure that headphones are not exposed to high voltage, they are often turned on as shown in Fig.1
dotted line, and a resistance of 4.7...10 kOhm is placed in the anode circuit. A transformer allows you to convert not only alternating voltage or current, but also the amount of resistance between the terminals of its windings. This explains the widespread use of transformers in low-frequency amplifiers. Let us assume, for simplicity of reasoning, that the efficiency of the transformer is 100%. Let's connect winding w1 of the step-down transformer Tr to the alternating current generator, and connect a load resistance of 100 Ohms to winding w2 (Fig.2)
. Fig.2
If the generator voltage is 100 V, and the transformation ratio n is equal to the ratio of the number of winding turns n = w1/w2 = 2, then the current I2 through the load resistance R2 and the power P2 in the load will be equal: I2 = U2/R2 = 50 V/100 Ohm = 0.5 A Since the efficiency of the transformer is 100%, the power in the load is equal to the power that the transformer consumes from the generator, that is, P1 = 25 W. The current in the generator circuit and winding w1 is equal to: I1 = P1/U1 = 25 W/100 V = 0.25 A.
Winding resistance w1 for generators is equal to: R1 = U1/I1 = 100 V/ 0.25 A = 400 Ohm.
Consequently, the resistance R1 turned out to be 4 times greater than R2. If we repeat the calculation for n = 3, we find that R1 will be 9 times greater than R2, etc. Therefore you can write: Thus, if resistance R2 is connected to one of the transformer windings, then the resistance of the other winding for the alternator is n times greater squared. If the transformer is a step-down transformer, then n is greater than one and the resistance R1 is greater than the resistance R2. For a step-up transformer, n is less than one and, as can be seen from formula (1), the resistance R1 is less than the resistance R2. Since resistance R1 depends only on the value of resistance R2, it is customary to say that R1 is the resistance reduced or recalculated to the primary winding. Using transformers with different transformation ratios, you can obtain a reduced resistance both greater and less than R2. On Fig.3
shows the most common circuit of a single-ended final stage on a beam tetrode (or pentode). Fig.3
The load of the lamp is the resistance of the loudspeaker Gr, converted into the primary winding w1 (but not the resistance of the winding w1!). As we have already indicated, the resistance of the voice coil of electrodynamic loudspeakers does not exceed 5...10 Ohms. Most electronic tubes designed to operate in the final stages of low-frequency amplifiers deliver maximum power at load resistance values of Ra 2.5...10 kOhm. The conversion of the low-resistance loudspeaker resistance R2p into the high-resistance load resistance Ra is carried out using an output transformer. It is easy to verify that the transformer must be a step-down transformer, and its transformation ratio can be found from formula (1). For real transformers, the efficiency is less than 100%. The required number of turns of the secondary winding w2, depending on the resistance of the loudspeaker voice coil, is found using the formula: where w1 is the number of turns of the primary winding specified in Table 1.
Table 1
Lamp type
6P1P
6P6S
6P14P
6P18P
6F1P*
6F3P*
Modes
Source voltage, V Output power, W** Reduced con. load, kOhm Automatic bias resistance, Ohm Anode current in rest mode, mA Output core cross-section trans., cm2 Number of turns of the primary winding Diameter of wire I winding, mm Diameter of wire II winding, mm * Pentode part of the lamp. In most circuits of final stages on beam tetrodes or pentodes, a capacitor Csh is connected parallel to the primary winding. Sometimes the capacitor Сш is connected between the anode of the lamp and the ground. As is known, the resistance of the voice coil of an electrodynamic loudspeaker largely depends on frequency and changes with frequency as shown in
Fig.4.
Fig.4
Approximately according to the same law, the resistance reduced to the primary winding, that is, the load resistance of the terminal lamp, changes with frequency. Changing the lamp load resistance leads to an increase in the nonlinear distortion coefficient. The resistance of a capacitor is known to decrease with increasing frequency. Therefore, a capacitor Csh is connected in parallel with the primary winding of the output transformer so that the lamp load resistance remains constant within the amplified frequency band. The capacitance of the capacitor Csh is chosen in the range from 3000 pF to 10000 pF. The operating voltage of the capacitor Ssh should be 2...3 times greater than the voltage of the anode power source. Typical values of resistance in the cathode circuit for terminal lamps and recommended modes of terminal lamps are given in table 1
. For lamps 6P1P, 6P6S, the rated power of this resistance must be at least 1 W, and for lamps 6P14P and 6P18P - at least 0.5 W. It is advisable to use resistances with a tolerance of +/- 5%. The capacitor Sk, which blocks the automatic bias resistance, must have a capacity of at least 10 µF for a 6P14P lamp and at least 5 µF for other lamps. For stable operation of the terminal lamps, the resistance Rc in the control grid circuit should not exceed 1 MOhm. Ultra linear amplifier
The main difference between an ultralinear amplifier ( Fig.5
) from the usual is that the shielding grid of the lamp is connected not to the plus of the power source, but to part of the turns of the primary winding of the output transformer. Fig.5
Constant voltage on shielding grids for circuits
Fig.3
And Fig.5
about the same. However, in the ultralinear amplifier circuit, the lamp’s shielding grid also receives an alternating output voltage, which is removed from the part of the primary winding between terminals 1-2. With the correct choice of lamp mode, non-linear distortions in the final stage are sharply reduced, and the output power and gain are reduced slightly. The frequency response of an amplifier with a transformer is determined mainly by the inductance of the primary winding L1 and the leakage inductance between the primary and secondary windings of the transformer. Fig.6
As you know, as the frequency decreases, the inductive reactance of the winding decreases, and therefore it will shunt the load resistance. And reducing the load resistance reduces the gain at lower frequencies. The lower the inductance of the primary winding L1 of the output transformer, the higher the frequency response of the amplifier begins to roll off (dashed curve in Fig.6
). In real output transformers, due to scattering, part of the magnetic lines of force created by the alternating current passing through the primary winding is closed, bypassing the turns of the secondary winding. This is the so-called leakage flux, which does not create an alternating voltage on the secondary winding. At low and medium frequencies this decrease is insignificant, but at the highest frequencies the voltage across the load decreases sharply. Conventionally, the effect of leakage flux can be imagined as some small inductance, the so-called leakage inductance Ls, connected in series with the primary winding of the output transformer. At low and medium frequencies, the value of the leakage inductance resistance is much less than the value of the recalculated load resistance. At the highest frequencies, this resistance increases and reduces the alternating voltage on the primary, and therefore on the secondary winding. The greater the leakage flux, the greater the leakage inductance and the worse the frequency response of the amplifier at higher frequencies (dashed line on
Fig.6
). Reducing leakage inductance is achieved by careful manufacturing of the output transformer and special design of the windings. In the simplest case, first half of the turns of the primary winding are wound, then the secondary and on top of it the remaining turns of the primary winding. The parts of the primary winding are connected in series, that is, the end of the first half to the beginning of the second. In single-ended output stages on lamps, a direct current always flows through the primary winding of the output transformer, which magnetizes the transformer core. This leads to two unpleasant phenomena. First, the undistorted output power of the amplifier is reduced. Therefore, for the same undistorted power, a transformer operating with permanent magnetization must have larger dimensions than a transformer without magnetization. Secondly, magnetization of the core by direct current causes a decrease in the magnetic permeability of the core material. This reduces the inductance of the primary winding of the output transformer, which in turn leads to a decrease in the cascade gain at the lowest frequencies, that is, to the appearance of frequency distortion. To weaken the influence of permanent magnetization, the core should be assembled with a gap of 0.1...0.2 mm between the W-shaped plates and jumpers. A paper gasket with a thickness of 0.1...0.15 mm is placed in this gap. Push-pull amplifiers
The schematic diagram of a push-pull triode amplifier is shown in
Fig.7
. Fig.7
From the diagram it can be seen that the constant component of the anode current of each lamp flows through half of the primary winding of the output transformer. The direction of the current in the halves of the windings is opposite and therefore the resulting magnetic field in the core turns out to be equal to the difference in the fields created by the current of each lamp. If the number of turns of the winding halves and the anode currents of the lamps are equal, the magnetic fields cancel each other out and the resulting magnetic field in the core is equal to zero. This is one of the important advantages of the push-pull circuit. The absence of magnetization of the core by direct current - constant bias - allows you to choose a core of smaller dimensions than for a single-ended one in amplifiers with the same output power. In addition, there is no need for clearance in the core. The grids of lamps L1 and L2 are supplied (usually from a phase inverter) with two voltages equal in amplitude but opposite in phase. Therefore, the anode currents of the lamps also change in antiphase, that is, when the anode current of one lamp increases, the anode current of the second lamp decreases ( Fig.8
). Fig.8
But since the halves of the primary winding of the output transformer are connected in opposite directions, the alternating magnetic field in the core turns out to be proportional to the arithmetic sum of the anode currents ( Fig.8c
). Therefore, the voltage on the secondary winding of the output transformer will be twice the voltage that it would be when operating one lamp. If each of the lamps of a push-pull circuit develops an output power Pout, then the total output power of the push-pull circuit will be equal to 2Pout. We could get the same power if we connected two lamps in parallel in a single-cycle circuit, but the push-pull circuit has a number of advantages, the most important of which are the absence of constant magnetization of the output transformer core; less nonlinear distortion due to the absence of even harmonics. Amplifier stages can operate in several modes, of which class A, B, AB, AB1, AB2 modes are used in LF amplifiers. Class A mode. The bias voltage on the control grids of the lamps - the operating point - of the class A amplifier is selected so that the alternating voltage of the signal on the grids of the lamps does not go beyond the straight section of the grid characteristic of the lamp ( Fig.9a
). Fig.9a
Amplifier performance in class A mode: low nonlinear distortion; the anode quiescent current of the lamp is greater than the alternating component of the anode current, due to which the efficiency is small and amounts to 30...40%. Class B mode. In class B mode, the operating point is selected at the lower bend of the grid characteristic of the lamps ( Fig.9b
). In this case, the anode quiescent current of the lamp is close to zero, so the anode current flows through the lamp only at positive half-waves of the input voltage. Class B mode is applicable only in push-pull circuits. In these circuits, the lamps in the arms operate alternately: during one half-cycle of the input voltage, the anode current passes through one lamp, and during the other half-cycle, through the other lamp. Fig.9b
AB class mode. Class AB mode occupies an intermediate position between modes A and B. The bias voltage on the control grid is chosen less than in a class B amplifier, but greater than in a class A amplifier ( Fig.9c
). As a result, the amplification of weak signals in this mode occurs in class A, and strong signals in class B. Nonlinear distortions in the AB mode amplifier are slightly higher than the distortions in mode A, and the efficiency is much more, especially at large amplitudes of the amplified signal. AB mode is used only in push-pull amplifiers. Fig.9c
AB mode amplifiers are divided into two groups: AB1, in which there are no grid currents, and AB2, in which work occurs with grid currents. Above we talked about various modes for amplifiers using vacuum tubes, but everything that has been said applies entirely to transistor amplifiers. Pre-amplification stages. A typical signal source used to develop an output voltage of 50-200 mV. High-quality amplifiers were oriented towards this voltage. Correction circuits were previously located between the input sockets and the grid of the first lamp, in which the signal was attenuated by at least half (6 dB) at the most sensitive input. In the fine-compensated volume control, the minimum signal attenuation is another 6 dB. Tone controls that provide ±20dB of control typically attenuate the signal by another 30-40dB. If there were cathode followers in the input circuits, the signal loss increased by another 3-6 dB. So, the total signal attenuation used to be 45-58 dB. The signal voltage on the grids of the final stage lamps averages 10-20 V. The ratio of this value to the input signal voltage is 10/0.05 = 200 (46 dB). So, the amplification of the preliminary stages, taking into account the signal attenuation and the required voltage on the grids of the final stage lamps, should previously have been on the order of 90-100 dB. In other words, the gain of the preliminary stages should be approximately 100,000. This is quite a significant value for a low-frequency amplifier. If the voltage gain of each of the amplifier stages is approximately 10, then, obviously, the number of stages should be equal to 5. If the gain of each stage is about 100, the total number of stages will be equal to 3 (with some margin). Since a gain of 10 per stage is provided by almost any modern low-frequency tube triode, and a gain of 100 per stage is the limit even for good low-frequency pentodes, it can be argued that for tube amplifiers the number of pre-amplification stages should range from three until five. How many cascades should you make: 3 or 5? The first answer, of course, is “3”. However, there is no need to rush. Three cascades - this means the minimum gain of the cascade is equal to the third root of 10000. Note that this is not the μ of the lamp, but the gain of the cascade, which rarely exceeds 50% of the μ of the lamp. Therefore, triodes are no longer needed. This means there will be three cascades on pentodes or, in extreme cases, two on pentodes and one on a triode. The latter circuit, which does not have any gain margin, does not allow the use of negative feedback in the circuit, i.e. practically unsuitable for Hi-Fi amplifiers, because without negative feedback it is impossible to reduce the coefficient of nonlinear distortion and expand the frequency range to the required values. Three stages on pentodes can allow the introduction of negative feedback, but then the first, input stage is also assembled on the pentode, and in this case, as experience shows, it is almost impossible to achieve a complete absence of microphone effect and a background level below 60 dB. The other extreme - five stages on triodes - always provides the required gain even on the worst tubes, however, using tubes with an average gain of about 20-50, it is easy to obtain the required gain with a sufficient margin with four triodes (i.e. on two double lamps). This scheme is the most common. True, many foreign companies produce a specially designed pentode for the input stage with a low level of self-noise and not prone to microphone effects (EF-184, EF-804, etc.). Using such a pentode and subsequent triodes with a large μ (90-120) of the ECC-83 type, it is possible to obtain the required gain on three stages using the pentode - triode - triode system, but firstly, such a system requires the use of special lamps, and - secondly - very high quality transformer steel, highly sensitive end lamps, etc. Therefore, this scheme is not suitable. Note. In the 21st century, the situation has changed significantly. Nowadays no one is using physical analogue pre-amplifier stages. Pre-processing of the signal is trusted to high-quality DACs. The input signal is considered normal at 1-2 volts. Therefore, for a tube terminal, an amplification of 20-50 times is sufficient. And this task is handled by one vacuum tube in the pre-amplifier stage. This is, for example, a double triode, which combines the functions of a bass reflex. That is why all the garbage from numerous successive cascades remains in the distant past. Evgeny Bortnik. Bass reflexes. If the phase inverter is assembled according to a circuit in which each arm is also an amplifier (for example, according to the circuit in Fig. 1), then the gain of this arm is taken into account in the overall gain of the path. We remind you that you need to take into account the gain of only one arm, since the second arm of the inverter is only a matcher for the second arm of the push-pull final stage and is not part of the general amplification path. If the phase inverter is assembled according to a symmetrical cathode follower circuit (Fig. 2), then its gain is always less than unity, so such a stage is not only not an amplification stage, but also requires an additional increase in the total gain by 4-6 dB. The method for selecting the gain for a transistor amplifier is exactly the same. Now specifically about the circuits of the pre-amplifier stages themselves. These are the simplest resistive amplifiers without any circuit features. Typical for all stages, both triodes and pentodes, are the anode (collector) loads reduced by 2-5 times compared to the optimal calculated values for expanding the bandwidth towards higher frequencies, increased to 0.1-0. 25 μF transition capacitors and up to 1-1.5 MΩ grid leakage resistors to reduce the frequency response rolloff at low frequencies, the use of negative current feedback in all stages except the one on which the frequency response control unit is assembled. As for the amplification elements themselves, in recent years many different new types of lamps and transistors with excellent parameters have appeared. Thus, the value of S for low-power lamps became equal to 30-50 mA/V against the usual values of 3-10 mA/V, and therefore the sensitivity of the lamps increased sharply. Calculations show that theoretically all the pre-amplification can be obtained even on two stages with such lamps. However, it would be useful to warn amateurs against haste in choosing such lamps. And the point here is not conservatism, but the fact that an increase in, say, the slope of the lamps is achieved by a sharp decrease in the gap between the control grid and the cathode, which significantly increases the tendency of the lamp to generate thermal currents and the resulting huge nonlinear distortions. Also important are the high cost and lower durability of such lamps. It can be argued that such tubes as 6N1P, 6N2P, 6NZP, 6N23P, 6N24P, 6Zh1P, 6Zh5P, proven by many years of practice, are quite suitable for the preliminary stages of even the best, most modern amplifiers. For example, below are shown several circuits of the CPU on lamps in their normal modes In Fig.3. tube pre-amp stages are shown. a - two-stage amplifier with interstage internal feedback; b - cascade with linearizing feedback in the protective grid circuit. Final and pre-final stages – power amplifiers. Formally, pre-terminal cascades (drivers, from the English word drive - excite, set, swing) are classified as voltage amplifiers, i.e., preliminary cascades, but they are discussed in this, and not in the previous paragraph, in order to emphasize that by the nature of the work and In terms of modes of use, drivers are much closer to final amplifiers, i.e. power amplifiers. Hi-Fi amplifiers are characterized by a significant output power of the order of 15-50W. This means that to excite (drive) the final stage without noticeable nonlinear distortions, a power of the order of 1-5 W is already required, at a voltage of up to 25-35 V, and if we take into account the requirements for reducing nonlinear distortions, it becomes clear that conventional low-power triodes cannot provide excitation of powerful terminal lamps. Therefore, it becomes logical and justified to use high-power lamps in the last voltage amplification stage. It is possible that, theoretically, it would be more correct to make the pre-terminal cascades in all cases transformer or choke in order to obtain the highest value of the anode voltage utilization factor ξ, but there are several reasons why this should not be done. The transformer cascade always introduces noticeable frequency distortions, and at powers above 1-2 W, noticeable nonlinear distortions. In addition, transformers are relatively expensive, complex and labor-intensive to manufacture, heavy and bulky, sensitive to magnetic interference and at the same time a source of audio frequency interference for other amplifier circuits (primarily input ones). At the same time, radio amateurs now have medium-power, broadband and economical lamps at their disposal, which make it possible to easily obtain undistorted power of about 2-4 W at an active load resistance. These primarily include lamps of types 6P15P, 6E5P, 6F3P, 6F4P, 6F5P, 6Zh5P, 6Zh9P, etc. However, this issue needs to be approached more carefully. In some cases, for reasons of simpler coordination, it is still advisable to use a transformer connection. Pre-amplifier circuits are shown below For final low-frequency cascades with a power of up to 10-12 W, radio amateurs in most cases use 6P14P type lamps, partly because they quite easily provide the specified power. In addition, unfortunately, there are no other lamps suitable for this purpose. Such an outdated, although very good lamp, like 6P3S (6L6) Nowadays It cannot be recommended, and the industry does not produce more powerful special lamps for ULF final stages like the German EL-34. [Strange conclusion, without any reason, in 1980-90 the use of 6P3S cannot be recommended! Pure voluntarism from the Soviet of Deputies. In the 21st century, for example, 6P3S lamps can be strongly recommended for designing a tube amplifier. It is important to find specimens in good preservation. E.B.] People often try to get more power from the same 6P14P tubes by forcing the mode, but this path is completely unacceptable due to the sharp deterioration in the reliability of the amplifier and the increase in nonlinear distortions when a grid thermal current appears. Taking into account the above, we can recommend that radio amateurs use 6P14P lamps in any push-pull circuits only at powers not exceeding 10 watts. [An amazingly meaningless recommendation in the style of “since there is nothing good, then do what you do.” The author seems to be a cool authority, but he writes nonsense. E.B.] With a higher output power, it is necessary to switch to such obviously not “low-frequency” lamps as 6P31S, 6P36S, 6P20S, GU-50, 6N13S (6N5S) both in classic push-pull and ultra-linear circuits, and in bridge circuits less familiar to radio amateurs circuits, also called push-pull-parallel. The first three of these lamps are intended for use in the final cascades of horizontal scanning televisions and allow you to extract power up to 25 W from two lamps; a GU-50 generator lamp with an anode voltage of 500-750 V (and according to its passport it has Ua.work = 1000 V) is easy delivers power of 40-60W in a push-pull circuit; double triode 6N13S, designed specifically as a control lamp in electronic voltage stabilizer circuits, has a very low internal resistance and, with a relatively low anode voltage, makes it possible to obtain a power of at least 15 W (per one cylinder) in a conventional push-pull circuit, and when switched on, two in each arm triodes in parallel (two cylinders) in conventional push-pull and bridge circuits provide an output power of up to 25 W. Using the listed lamps, the radio amateur has a wide choice for creative activities. [Another recommendation in a vague state of consciousness. I wonder why twin or triple lamps are not suitable for creative activities? Maybe the author simply does not know the rules for parallel connection of radioelements? Namely, a parallel connection, with a high-quality selection of copies, gives a lot of intermediate options for very powerful amplifiers with decent characteristics. It is strange to read the recommendation of a 6P31S lamp, which is not at all more powerful than 6P14P, but is much worse in characteristics. And it’s also disappointing to see quick recommendations for the use of 6N13C lamps (paralleled, by the way). An amazing demonstration of frivolity, since the author is completely unaware of practice, because 6N13C lamps are rare guano. The spread of characteristics of the halves has a range of 100% or more. It is almost impossible to accurately select them for parallel connection, so the amplifier cannot deliver significant power to the load without overheating one of the halves, and the utilization factor is unlikely to exceed 40-50%. And simple parallel circuits for 6N13S, without leveling body kits, are unsuitable. And the discussions about lamps are touching, because there are a large number of other excellent lamps, in contrast to the recommended ones, for example 6P13S, 6P44S, 6P45S, G807; in extreme cases, 6P3S lamps are suitable. E.B.] Fig.5. Powerful final stages of the low-frequency ULF path. a - on 6P36S lamps in ultralinear switching; b - on GU-50 lamps in a push-pull parallel circuit; c - on 6N13S lamps with fixed bias balancing Since all circuits were considered as low-frequency, i.e. designed for a limited bandwidth (no more than 5-8 kHz), nothing was said about output transformers, chokes, and autotransformers. All of them are the most common, assembled on W-shaped or strip cores made of simple transformer steel 0.35 mm thick. There are no increased requirements for the frame design and windings, with the exception of a high degree of symmetry of the individual halves of the primary winding. This requirement is especially important for ultra-linear circuits for switching terminal lamps. The values of leakage inductance and capacitance of the primary winding are not significant. Secondary windings with powers above 10 W should be wound with as thick a wire as possible to reduce active losses. It is advisable to make several taps to select the best operating mode for the final stage. This issue is discussed in more detail in the next paragraph. The high-frequency final stages of two-channel Hi-Fi amplifiers are significantly different from the low-frequency ones, so the recommendations regarding them will be different. First of all, this applies to the types of lamps. [ Amazing reasoning. The author invented his own classification of LF and HF. Even to a complete amateur who has read the section on vacuum tubes, first of all, it is obvious that the invented frequency division has nothing to do with vacuum tubes at all; their range goes into hundreds of megahertz. The 6P14P lamp is purple, which frequency signals should be amplified, be it 0.1 kHz, 1 kHz, 5 kHz, 8 kHz, 16 kHz or 32 kHz. But with regard to the matching transformer, this question is already relevant. But there is no need to worry here either, because... up to 18-20 kHz, ordinary transformers are suitable; you don’t need to wind anything at all. And for frequencies above 20 kHz you should switch to ferrites. It seems that the author has not heard anything about sectioning the windings to improve the frequency response, and recommends a thick wire for the secondary winding. And the concept of ACTIVE LOSSES is absolute bullshit, since there are no passive losses and there are no reactive losses either. E.B.] Since the power of high-frequency channels, even in top-class amplifiers, is in the range of 10-12 W, the most suitable lamps are 6P14P and 6N13S. The best switching circuits are push-pull ultralinear, bridged on 6P14P in triode switching, and “two-story” on 6N13S. Regarding the last scheme, the most common version of which is shown in Fig. 6, we can say that although it is not new in a theoretical sense, it became widespread in broadcasting equipment only in the 60s of the last century. As often happens, the scheme has become very widespread, and when talking about the advantages of the scheme, they usually remain silent about its disadvantages. Let's try to objectively evaluate both. [First of all, I propose to sensibly evaluate the most important consequence of the creation of transformerless circuits. The past 50 years have shown that such schemes have not received any distribution, and could not have received them. As the standard of living rises, the value of health increases. Therefore, the main and insurmountable disadvantage of transformerless circuits - the lack of galvanic isolation from a high voltage source - will never allow such circuits to achieve at least some distribution among the human population. And let the dreamers study and analyze the modes of such circuitry until they are blue in the face.] Fig.6. One of the most common final stage circuits with series connection of DC lamps Connecting two lamps in series for direct current is equivalent to the fact that for alternating current both of them are connected in parallel relative to the load, due to which their total internal resistance is actually four times less than that of a conventional push-pull cascade. If for such a circuit we take lamps whose internal resistance is lower than usual, and use relatively high-impedance loudspeakers as a load, then it turns out that the output transformer, according to calculations, would in this case have a transformation coefficient close to unity or, in any case, measured in units. It is then possible to connect the load to the lamps directly, without an output transformer. This, of course, is an unconditional advantage of the scheme. However, this dignity comes at a high price. First of all, direct switching on of the load still turns out to be impossible due to the presence at the points of its switching on of half the voltage of the power source (120-150V). Therefore, the loudspeakers have to be switched on through a decoupling capacitor, the capacitance of which is directly related to the load resistance and the lower limit of the passband. Indeed, if the permissible voltage loss of the useful signal on the separating capacitor is 10% of the value of the signal itself, then at Rн=20 Ohm and flow=40 Hz the reactance of the capacitor should not exceed 2 Ohms, from which its capacitance is equal to It is clear that only an electrolytic capacitor can have such a capacitance, but it must be remembered that its operating voltage must be at least not lower than the full voltage of the power source, i.e. 300-350V. And then it turns out that the cost of such a capacitor is not at all lower than the cost of the output transformer, especially since, unlike a capacitor, a radio amateur can always make a transformer himself, if necessary. Of course, it is possible to make a loudspeaker with a voice coil resistance of not 20, but 200 Ohms, which, under the same conditions, will allow reducing the capacitance of the coupling capacitor to 200 μF, but in this case the cost of the loudspeaker increases sharply. However, this is not the only drawback of this scheme. The second is that when the lamps are connected in series with direct current, only half the voltage of the anode source is applied to each of them, so the circuit can only work well on special lamps whose rated anode voltage does not exceed 100-150V. However, most lamps of this type have an insignificant maximum output power, rarely exceeding a few watts. In addition, studies have shown that when using pentodes, this circuit is fundamentally somewhat asymmetrical, which makes it unsuitable for the final low-frequency stages of Hi-Fi amplifiers. In high-frequency cascades, the first drawback immediately disappears, since with the values selected in the previous calculation and the lower limit of the HF channel flow = 2 kHz, the capacitance value of the separating capacitor Moreover, in this case, a ten percent signal loss will occur only in the worst, practically non-working part of the passband, and at ftop = 20 kHz the signal loss will be only 1%. In addition, the required output power for the final RF stage is significantly less than for the LF stage, which allows the use of a 6N13C double triode in this circuit, which has low internal resistance and works well at low anode voltages. A practical diagram of such a cascade is shown in Fig. 7. Fig.7. Practical diagram of a “two-story” final stage based on a double triode 6N13S (6N5S) If the power of the RF channel does not exceed 2-3W, you can assemble the final stage according to the circuit in Fig. 8 using lamps of types 6F3P or 6F5P. The output transformer for this circuit is assembled on a tape core with a tape thickness of no more than 0.2 mm or on an W-shaped permalloy. In order for the ultralinear circuit to give a noticeable result and for nonlinear distortions to actually be on the order of 0.2-0.5%, the tap point of the primary winding must in each case be selected empirically directly from the results of measurements of the r.n.i. in the process of setting up an amplifier. To do this, when winding a transformer, 4-6 taps must be provided for each half of the primary winding. Fig.8. Push-pull high-frequency final stage using 6F3P or 6F5P lamps (Pout = 2.5 W) For transistor amplifiers, the “two-story” circuit, on the contrary, turns out to be preferable to all others. This is explained by the low internal resistance of high-power transistors and collector voltage (compared to lamps). Therefore, excellent matching of the cascade with the load is ensured even when using conventional low-impedance loudspeakers, for example, the 4GD-35 type. In addition, the decoupling capacitor turns out to be small in size even with a capacity of 2000-5000 μF, since its operating voltage does not exceed 20-30V. Such schemes are widespread and well known to radio amateurs. As a general conclusion, I can cite several considerations that in the 21st century will certainly be perceived as rational. The first consideration is whether it is correct for the author to discuss only push-pull amplifiers, since single-ended circuits are intended for beginners. Secondly, the thoroughness of the approach to systematizing the circuitry of cascades also deserves respect. Third, the author’s indisputable qualifications in some cases border on astounding prejudices, and lapses in thinking are apparently a consequence of the author’s high theoretical preparation and insufficient practical experience. Fourth, the past decades have significantly changed the situation, both in basic concepts and in circuit design, especially with regard to the output stages of high-performance amplifiers. And there is no longer any excessive ceremony. Much has become simpler and clearer. Some show-offs died without showing resilience. But they will be replaced by new show-offs, like oxygen-free copper. It seems very important to understand the fact that changes in the technological structure of society should not change the fundamental values of life, for example, the Slavic civilization. Prepared a publication based on materials from Gendin’s book downloaded online. Evgeny Bortnik, Krasnoyarsk, Russia, March 2018 Low-frequency amplifiers are mainly designed to provide a given power to the output device, which can be a loudspeaker, a tape recorder's recording head, a relay winding, a measuring instrument coil, etc. The input signal sources are a sound pickup, a photocell, and various converters of non-electrical quantities into electric. As a rule, the input signal is very small, its value is insufficient for normal operation of the amplifier. In this regard, one or more pre-amplifier stages are included in front of the power amplifier, performing the functions of voltage amplifiers. In ULF preliminary stages, resistors are most often used as a load; they are assembled using both lamps and transistors. Amplifiers based on bipolar transistors are usually assembled using a common emitter circuit. Let's consider the operation of such a cascade (Fig. 26). Sine wave voltage u in supplied to the base-emitter section through an isolation capacitor C p1, which creates a ripple of the base current relative to the constant component I b0. Meaning I b0 determined by source voltage E k and resistor resistance R b. A change in the base current causes a corresponding change in the collector current passing through the load resistance R n. The alternating component of the collector current creates at the load resistance Rk amplitude-amplified voltage drop u out. The calculation of such a cascade can be done graphically using those shown in Fig. 27 input and output characteristics of a transistor connected according to a circuit with an OE. If load resistance R n and source voltage E k are given, then the position of the load line is determined by the points WITH And D. At the same time, the point D given by value E k, and point WITH– electric shock I to =E k/R n. Load line CD crosses the family of output characteristics. We select the working area on the load line so that signal distortion during amplification is minimal. For this, the intersection points of the line CD with output characteristics must be within the straight sections of the latter. The site meets this requirement AB load lines. The operating point for a sinusoidal input signal is in the middle of this section - point ABOUT. The projection of the segment AO onto the ordinate axis determines the amplitude of the collector current, and the projection of the same segment onto the abscissa axis determines the amplitude of the variable component of the collector voltage. Operating point O determines the collector current I k0 and collector voltage U ke0 corresponding to the rest mode. Moreover, point O determines the base quiescent current I b0, and therefore the position of the operating point O" on the input characteristic (Fig. 27, a, b). To points A And IN output characteristics correspond to points A" And IN" on the input characteristic. Line segment projection A"O" the x-axis determines the amplitude of the input signal U in t, at which the mode of minimal distortion will be ensured. Strictly speaking, U in t, must be determined by the family of input characteristics. But since the input characteristics at different voltage values U ke, differ slightly, in practice they use the input characteristic corresponding to the average value U ke=U ke 0.
The simplest circuit of the final stage - a circuit with direct connection of the load - is shown in Fig.1
.
The most common load for the final stages of radio broadcast receivers is an electrodynamic loudspeaker with a voice coil resistance of 3...10 Ohms. Such loudspeakers are included in the anode circuits of the final stages through an output transformer. Currently, electrodynamic loudspeakers with a resistance of 200...800 Ohms have been developed, which can be connected to an amplifier without output transformers.
P2 = U2 I2 = 50 V x 0.5 A = 25 W.
(1)
(2)
** The output power is indicated taking into account losses in the output transformer.
The inductance of the primary winding of the output transformer is chosen such that the inductive resistance of this winding is greater than the resistance of the loudspeaker converted into the primary winding. This is easily done at medium audio frequencies, at which the frequency response of the cascade is uniform ( Fig.6
).
The advantage of class B mode is its high efficiency. - up to 60...75%. It should be borne in mind that for mode B amplifiers it is impossible to create a bias on the lamp grids using resistances in the cathode circuit.
.png)
.png)
