thesis

2.1 Choosing a LNA Circuit

In accordance with the above considerations, it is necessary that the low noise amplifier meet the following specifications:

gain factor not less than 20 dB;

noise figure no more than 3 dB;

dynamic range not less than 90 dB,

center frequency 808 MHz.

in addition, it had high stability of characteristics, high reliability of operation, small dimensions and weight.

Taking into account the requirements for a low-noise amplifier, we will consider possible options for solving the problem. When considering possible options, we will take into account the conditions under which the receiving-transmitting module will be operated (placement on board the aircraft and the impact of external factors such as temperature differences, vibrations, pressure, etc.). Let us analyze low-noise amplifiers made using different element base.

The quietest of microwave amplifiers are currently quantum paramagnetic amplifiers (masers), which are characterized by extremely low noise temperatures (less than 20 o K) and, as a result, very high sensitivity. However, the quantum amplifier includes a cryogenic cooling system (up to a liquid helium temperature of 4.2 o K), which has large dimensions and weight, high cost, as well as a bulky magnetic system for creating a strong constant magnetic field. All this limits the scope of quantum amplifiers to unique radio systems - space communications, long-range radar, etc.

The need to miniaturize microwave radio receivers, increase their efficiency, and reduce costs has led to the intensive use of low-noise amplifiers based on semiconductor devices, which include semiconductor parametric, tunnel diode, and microwave transistor amplifiers.

Semiconductor parametric amplifiers (SPA) operate in a wide frequency range (0.3 ... 35 GHz), have bandwidths from fractions to several percent of the center frequency (typical values ​​\u200b\u200bof 0.5 ... 7%, but bandwidths up to 40% can be obtained); the transmission coefficient of one stage reaches 17…30dB, the dynamic range of input signals is 70…80dB. As pump generators, generators based on avalanche-span diodes and Gunn diodes, as well as on microwave transistors (with and without frequency multiplication) are used. Semiconductor parametric amplifiers are the most low-noise of semiconductor and, in general, of all uncooled microwave amplifiers. Their noise temperature ranges from tens (at decimeter wavelengths) to hundreds (at centimeter wavelengths) of degrees Kelvin. With deep cooling (up to 20 o K and below), they are comparable in noise properties to quantum amplifiers. However, the cooling system increases the dimensions, weight, power consumption and cost of the PPU. Therefore, cooled PPUs are used mainly in terrestrial radio systems, where highly sensitive radio receivers are required, and the dimensions, weight, and power consumption are not so significant.

The advantages of PPU in comparison with amplifiers based on tunnel diodes and microwave transistors, in addition to better noise properties, include the ability to operate in the higher frequency range, greater amplification of one stage, the possibility of quick and simple electronic frequency tuning (within 2 ... 30%). The disadvantages of PPU are the presence of a microwave pump generator, a smaller bandwidth, large dimensions and weight, and a significantly higher cost, in contrast to transistor microwave amplifiers.

Compared to other semiconductor amplifiers, tunnel diode amplifiers have smaller dimensions and weight, determined mainly by the dimensions and mass of ferrite circulators and valves, lower power consumption and a wide bandwidth. They operate in the frequency range 1 ... 20 GHz, have a relative bandwidth of 1.7 ... 65% (typical values ​​​​3.5 ... 18%), the transmission coefficient of one stage is 6 ... 20 dB, the noise figure is 3.5 ... 4.5 dB at decimeter waves and 4 ... 7 dB on centimeter, the dynamic range of input signals is 50 ... 90 dB. Tunnel diode amplifiers are mainly used in devices where it is necessary to place a large number of light and small-sized amplifiers in a small area, for example, in active phased antenna arrays. However, due to their inherent disadvantages (relatively high noise figure, insufficient dynamic range, low dielectric strength of the tunnel diode, difficulty in ensuring stability, the need for decoupling devices), amplifiers based on tunnel diodes have recently been intensively replaced by microwave transistor amplifiers due to their inherent disadvantages.

The main advantages of semiconductor low-noise amplifiers - small size and weight, low power consumption, long service life, the ability to build microwave integrated circuits - allow them to be used in active phased antenna arrays and in on-board equipment. Moreover, microwave transistor amplifiers have the greatest prospects.

Advances in the development of semiconductor physics and technology have made it possible to create transistors with good noise and amplifying properties and capable of operating in the microwave range. Based on these transistors, microwave low-noise amplifiers were developed.

Transistor amplifiers, unlike amplifiers based on semiconductor parametric and tunnel diodes, are not regenerative, so it is much easier to ensure their stable operation than, for example, amplifiers based on tunnel diodes.

Microwave LNA uses low-noise transistors, both bipolar (germanium and silicon) and field-effect transistors with a Schottky barrier (based on silicon and gallium arsenide). Germanium bipolar transistors allow you to get a lower noise figure than silicon, but the latter are more high-frequency. FETs with a Schottky barrier are superior to bipolar transistors in amplifying properties and can operate at higher frequencies, especially gallium arsenide transistors. Noise characteristics at relatively low frequencies are better with bipolar transistors, and at higher frequencies with field ones. The disadvantage of field-effect transistors is their high input and output impedance, which makes broadband matching difficult.

The above considerations allow us to outline a strategy for the synthesis of a low-noise field-effect transistor amplifier in a monolithic integrated design.

As was chosen earlier, the LNA will be built on the basis of the MGA - 86563 module. The electrical circuit diagram is shown in Figure 2.1. A typical switching circuit is shown in Figure 2.2: Figure 2.1 Electrical circuit diagram MGA-86563. Figure 2...

High frequency receiving path

As a result of the work carried out, the MGA86563 low-noise amplifier was investigated. The study of the frequency response of the LNA was carried out using the stand SNPU-135, a device for studying the frequency response X1-42. The connection diagram for measuring the frequency response is shown in Figure 4...

AC to DC voltage measuring converter

To implement the rectifier circuit, we use a dual high-speed op-amp with field-effect transistors at the input of the KR140UD282 type. Its parameters are given in Table 5, and the switching circuit is shown in Fig. 8...

Low noise integrated amplifier

Modeling in the MICRO-CAP system of measuring transducers based on temperature sensors

Based on the building, it is necessary to build a three-wire circuit (2 options) for temperature measurement using RTD using a current source (see Figure 6.2.1). No. Scheme Voltage at the input of the DUT at 2 Fig.6.2.1...

Designing the amplifying part of the device

Let's use the scheme shown in fig. 5 to calculate the power amplifier. When calculating the MA, the given values ​​are: a). Rated power in the load Рн = 0.4 W; b). Load resistance Rн = 100 Ohm...

The process of modeling the operation of the switching node

Since the common mode noise does not exceed 10V and the gain is not large, it will be enough to take the simplest differential amplifier. The circuit of the simplest differential amplifier is shown in Figure 5 ...

Transducer Development

Figure 2 The preamplifier (PA) is an operational amplifier (op amp) with negative feedback. The switching circuit (PU) is shown in Figure 2 ...

Calculation of a pulse amplifier

The pulse voltage amplifier is a signal pre-amplifier that ensures the normal operation of the PA...

Synthesis of an inverting amplifier

Diagram of an inverting amplifier with negative feedback: Figure 1 - Basic circuit of an inverting op amp with OOS ...

For the convenience of developing and performing calculations, the blocks PU, ULF and UHF2 were combined into a common scheme. The construction was based on the 140-UD20A microcircuit and KT817A bipolar transistors ...

Comparative characteristics of the technical data of radio stations

Figure 7.5 shows the electrical circuit diagram of the UHF2 preamplifier, low frequency amplifier and high frequency amplifier. The circuit is based on the 140-UD20A microcircuit, which consists of operational amplifiers (Da1 ...

Microphone amplifier circuit

Let's determine the total gain, based on which the number of amplifying stages is selected where is the total gain; effective rated output voltage; effective rated input voltage...

Broadband amplifier

Starting the development of an amplifier, it is necessary to be guided by general considerations of the economic feasibility of its production (minimization of active devices, elements and components by their number ...

There are many amplifiers for which one of the main required parameters is the requirement to ensure minimal output noise. Typically, such circuits are used to amplify signals from various sensors, as well as in direct conversion receivers, where the main amplification is carried out at low frequencies. An increase in noise makes it impossible to distinguish weak signals against a background of noise.

The internal noise of the amplifier occurs when current passes through the passive and active elements of the circuit.
Noise characteristics also depend to a large extent on the construction of the circuit (circuitry). When developing an amplifier with a high signal-to-noise ratio, in addition to the optimal choice of the type of circuit, it is important to choose the right element base and optimize the operation of the cascades.

Selecting Schematic Components

In a real amplifier, the source of internal noise is:
1) thermal and current noise of resistors;
2) flicker noise of capacitors, diodes and zener diodes;
3) fluctuation noise of active elements (transistors);
4) vibration and contact noise.

Resistors

The inherent noise of resistors is the sum of thermal and current noise.

Thermal noise is caused by the movement of electrons in the conductive material from which the resistor is made (this noise increases with increasing temperature). If no voltage acts on the resistor, then the noise EMF on it (in μV) is determined from the relation:

Esh=0.0125 x f x R,
where f is the frequency band in kHz; R is the resistance in kOhm.

Current noise occurs when current flows through a resistor. In this case, the noise voltage appears due to the effect of fluctuations in the contact resistances between the conductive particles of the material. Its value linearly depends on the applied voltage. Therefore, the noise properties of resistors are characterized by the noise level, which is the ratio of the effective value of the variable component of the noise voltage Em (μV) to the applied voltage U (V): Em / U.

The frequency spectrum of both types of noise is continuous ("white noise"). And if for thermal noise it is evenly distributed up to very high frequencies, then for current noise it begins to fall already from about 10 MHz.

The total amount of noise is proportional to the square root of the resistance, so to reduce it, the resistance value in the circuit must also be reduced.
Sometimes, in order to reduce the noise caused by resistors, they resort to parallel (or series) connection, and also set more power than is required for operation. In addition, it is possible to use those types in which, due to the manufacturing technology, this parameter is less.

Non-wire resistors have much more current noise than thermal noise. The overall noise level for different types of resistors can range from 0.1 to 100 µV/V.

To compare various resistors (fixed and trimmers from the SP group), the maximum noise values ​​\u200b\u200bare shown in table 1

Type of resistors Technological version Noise level, µV/V BLT brown-carbon 0.5 S2-13 S2-29V metal-dielectric 1.0 S2-50 metal-dielectric 1.5 MLT OMLT S2-23S2-33 metal-dielectric 1...5 S2-26 metal oxide 0 .5 SP3-4
SP3-19
SP3-23 film composite 47...100
25...47
25...47
Table 1 - Noise properties of resistors

As can be seen from the table, tuned resistors are much more noisy. For this reason, it is better to use them with small denominations or to exclude them from the circuit altogether.
The noise properties of resistors can be used to implement a broadband noise generator.

As recommendations for choosing resistors for assembling a low-noise amplifier, it can be noted that it is most convenient to use the types: C2-26, C2-29V, C2-33 and C1-4 (unpackaged chip design). Recently, low-noise imported metal-dielectric resistors have appeared on sale, similar in design to C2-23, but with a lower noise figure (0.2 μV / V).

It is possible to significantly reduce the noise of resistors by strongly cooling them, but this method is too expensive and is used very rarely.

Capacitors

In capacitors, the source of flicker noise is the leakage current. Large-capacity oxide capacitors have the highest leakage currents. Moreover, the leakage increases with an increase in capacitance and decreases with an increase in the allowable rated operating voltage.

Reference data for the most common oxide capacitors are given in table 29.
The smallest leakage currents among polar capacitors are: K53-1A, K53-18, K53-16, K52-18, K53-4 and others.
Oxide capacitors installed at the input as isolation capacitors can significantly increase the noise of the amplifier. Therefore, it is desirable to avoid their use, replacing them with film ones (K10-17, K73-9, K73-17, KM-6, etc.), although this will lead to a significant increase in the size of the structure.

Capacitor type Manufacturing technology Operating temperature, C Leakage current, µA K50-6
K50-16
K50-24
aluminum oxide-electrolytic -10...+85
-20...+70
-25...+70 4...5000
4...5000
18...3200 K52-1
K52-2
K52-18 tantalum oxide volume-porous -60...+85
-50...+155
-60...+155 1,2...8,5
2...30
1...30 K53-1
K53-1A
K53-18 tantalum oxide-semiconductor -80...+85
-60...+125
-60...+125 2...5
1...8
1...63
Table 2 - Reference parameters of capacitors

Diodes and zener diodes

With direct current flow, the noise of the diodes is minimal. The greatest noise is provided by the leakage current (under the action of reverse voltage), and the smaller it is, the better. Quite a lot of noise from zener diodes. This property is even sometimes used to perform the simplest noise generators for children's toys (simulators of surf noise, fire sounds, etc. - L16, L17). To obtain maximum noise in such circuits, zener diodes operate at low currents (with a large additional resistor).

transistors

In the transistor itself, the main types of noise are thermal and generation-recombination noise, the power spectral density of which does not depend on frequency.

To reduce the noise level, low-noise bipolar transistors with a normalized noise figure (Ksh) are usually used in our country to work in the input stages. These are: (p-p-p) KT3102D (E), KT342V and (pn-p) KT3107E (W, L) and a number of others. It should be noted here that the use of low-noise high-frequency bipolar transistors in the low frequency range, as a rule, happens to be inappropriate. For such transistors, the noise figure is normalized only in the high-frequency region, and in the range below 100 kHz, they can make noise no less than any others. In addition, such transistors may exhibit a tendency to excitation (self-generation).

If it is necessary to obtain a large input impedance in the input stage of the amplifier, the KP303V(A) field effect transistor is often used. It is made with a p-n junction gate (n-type channel) and has a rated noise figure.

contact noise

occur when poor-quality soldering (with violation of the temperature regime) or at the junction of the connectors. For this reason, it is not recommended to connect the input circuits of a low-noise amplifier through plug-in connections. I have also come across a situation where transistors after re-soldering made more noise in the same circuit.

Vibration noises

may occur when the device is operated on moving objects or in places with increased vibration from operating equipment. They arise due to the transmission of mechanical vibrations to the capacitor plates, between which there is a potential difference (the so-called "piezo-microphone effect"). This is observed even in small-sized ceramic capacitors (K10, K15, etc.) of increased capacitance (more than 0.01 μF). Such interference can be especially strong in coupling capacitors installed at the input of the amplifier. The interference signal during mechanical vibrations has the form of short pointed pulses, the spectrum of which is in the low frequency range. To combat this type of interference, you can apply depreciation of the entire structure. In oxide capacitors, these interferences do not occur.

When choosing parts for assembling a low-noise circuit, it is necessary to take into account their production time. The manufacturer guarantees the parameters only for a certain period of storage. This is usually no more than 8 ... 15 years. Over time, aging processes occur, manifested in a decrease in insulation resistance, the capacitance of capacitors decreases and leakage currents increase. Oxide capacitors change their characteristics especially strongly over time. For this reason, it is best to avoid their use in signal chains whenever possible.

V. P. Matyushkin, Drogobych

The features of the spectrum of nonlinear distortions in amplifiers with different cutoff frequencies are compared. It is shown that devices based on operational amplifiers enrich the audio signal with higher harmonics, so their use in audio complexes of especially high quality is undesirable. The design of a low-noise, highly linear preamplifier with a high cutoff frequency and volume and tone controls is presented.

When using passive tone controls (RT) and sufficient sensitivity of the UMZCH, the purpose of the pre-amplifier ZCH (PUZCH) remains to compensate for the attenuation of the amplified signal introduced by the RT and to match the input and output impedances of the various links of the path with each other. This function belongs to linear low-noise amplification stages with high (tens to hundreds of kΩ) input and low (no more than 600 Ω) output impedance. Such values ​​are necessary so that errors are not introduced into the characteristics of the regulation of the RT and the volume control (RG) and do not affect the characteristics of the signal sources.

The PUZCH designs known to the author do not meet the increased requirements for them. If earlier, when playing a gramophone or tape recording, it was quite enough that the relative noise level of the BUZCH was about -80 ... -85 dB, which is no worse than that of signal sources, then when listening to CDs, when "dead silence" in pauses is filled with an annoying hiss, such noise is already becoming an annoying hindrance. Other parameters also leave much to be desired, especially for PUZCH made using operational amplifiers (op-amps).

The low (tens to hundreds of hertz) op-amp cutoff frequency fc causes not the best transient response, which determines the fidelity of the pulse signal edge transmission. Such fc forces one to take into account the possibility of dynamic distortions, and also leads to a decrease in the depth of the FOS with increasing frequency, i.e. to the growth of non-linear distortions (NI). The deterioration of the suppression of signal distortions begins in the op-amp covered by the OOS, from its cutoff frequency to and occurs approximately in direct proportion to the frequency. For example, if fc<500 Гц и при усилении сигнала с частотой fA=1 кГц получен уровень второй гармоники (на частоте 2 кГц) 0, 001%, то при усилении равного по амплитуде сигнала с частотой fB=8 кГц уровень второй гармоники (на частоте 16 кГц) будет примерно в fB/fA=8 раз больше, что дает уже не такие благополучные искажения (0, 008%). Однако это еще только полбеды.

Even worse is that at the same time the ratio between the harmonics of the same signal changes in favor of harmonics of a higher order. This applies to NIs generated by those cascades of op amps (first of all, output ones, due to the significance of their contribution to the overall NI level), which follow the cascade that forms a break in the frequency response at the frequency fc. The distortions of these cascades will be kept in mind further (in the first cascades of the op-amp, the processes have their own characteristics).

Figure 1 shows the frequency dependences of the ratio of the coefficient of NI for the harmonic n>2 Qn to the coefficient of NI for the second harmonic Q2, reduced to the same ratio for the OS without FOS Qn/Q2. Line 1 corresponds to OS without OOS, line 2 - OS with closed loop OOS. Line 1 also corresponds to an amplifier that has a high cutoff frequency fc ">> 20 kHz, and it does not matter whether the feedback is turned on or not. As you can see, the ultrasonic frequency amplifier on the op amp enriches the NI spectrum with higher-order harmonics. The observed picture is smoothed only by the fact that the original (without feedback ) the amplitudes of the harmonics themselves usually decrease with an increase in their number n, so the distortion products recorded during measurements do not depend so much on the frequency.It is clear that a picture similar to Fig. 1 also occurs for intermodulation distortion components of various orders.

As you know, the sound quality depends not only on the amplitudes of harmonics of different orders, but also on the ratio between them: it is desirable that with an increase in the harmonic number, its amplitude decreases rather quickly, otherwise the sound becomes hard, acquires an unpleasant metallic tint. It can be seen from Fig. 1 that the UZCH on the op-amp acts in the opposite direction, and in almost the entire sound range, excluding only the lowest frequencies (and this applies, of course, not only to the PUZCH, but also to power amplifiers). And if the bass tone control, raising the frequency response of the path at frequencies below 1 kHz, to some extent restores the ratio between the harmonics in the slope range of its frequency response section, then raising the high frequencies by the high frequency tone control further exacerbates the violation of the ratio between them at frequencies greater than 1 kHz.

Thus, the notorious "transistor sound" begins to emerge even in the PUZCH, made on the op-amp. Therefore, the enthusiasm for such schemes, despite all the convenience and simplification when using the op-amp, is at the expense of the quality of sound reproduction. And there is nothing surprising in the fact that they sound worse than tube amplifiers, which, as a rule, have a fairly high fc (which is possible due to relatively shallow feedback) and, moreover, a favorable spectrum of harmonics generated by tubes (no higher than the fifth order).

To obtain a favorable NI spectrum, the transistor amplifier must have a cutoff frequency fc "\u003e 20 kHz (Fig. 2, curve 1) before covering the OOS (Fig. 2, curve 1). This requirement is also well consistent with the condition of the absence of dynamic distortions. At the same time, the possibility of additional improvement of the spectrum of harmonics and approximation of its character to the lamp one by specific correction, which consists in raising the original (without OOS) frequency response with increasing frequency in the audio range or at least in some of its section (Fig. 2, broken line 3).Curve 2 corresponds to case 2 of Fig. 1. Due to a decrease in the relative proportion of high-frequency components in the NI, this would make it possible to obtain a distortion spectrum in Fig. 1, curve 3, which should, apparently, make the sound softer.However, this issue still needs to be studied.

The disadvantages of the known PUZCH become especially noticeable when working together with modern high-quality UMZCH, for example.

When developing the proposed PUZCH, the above considerations were taken into account; at the same time, it is desirable to achieve the maximum simplicity of the circuit.

Amplifier parameters (Fig. 3):
Cutoff frequency fc 300 kHz
Intermodulation NI coefficient at 11out< 5 В и Rh >1 kΩ in the range of 0.02-20 kHz< 0, 001 %
Rated Iin 0.25 V
Maximum I out 9V
Noise level (R^0) -103 dB
Weighted value -109 dBA
output impedance< 0, 1Ом
Phase angle at f=0, 1 ...200 kHz< 0, 1°
Minimum load resistance R 300 Ohm

The amplifier is made according to a symmetrical circuit on complementary pairs of transistors, such a structure significantly increases its initial linearity even before the OOS coverage. All transistors, including output ones, operate in class "A" mode, and the collector quiescent current VT7, VT8 is about 10 mA and allows them to maintain this mode at load resistances Rh of at least 300 ohms.

Despite the fact that VT5 and VT6 are connected according to a common emitter circuit, their transfer characteristics are quite linearized by significant resistances in the emitter circuits (R15, R16).

The NI level turned out to be so low that it was decided not to use the envisaged EPOS loops, which would significantly complicate the scheme.

The input stage, in order to obtain a low noise level, is made on field-effect transistors with a pn junction. The input impedance of the amplifier, equal to about 350 kOhm, is determined only by the resistances of the resistors R3, R6 (in this case, one should not forget about the corresponding change in the capacitances C1, C2 so that the time constants of the HPF R3C1 and R6C2 remain the same). The voltage dividers R1R2 and R4R5R7 set the operating points VT1 and VT2, the resistor R4 serves to initially set zero voltage at the amplifier output and after tuning it can be replaced with a constant resistor of the desired resistance, and the value of the constant component at the amplifier output is not so critical and can be within ± 200 mV.

To obtain a large gain of the input stage and low noise, a dynamic load was applied on field-effect transistors VT3, VT4. Since both arms of the input stage (VT1-VT3 and VT2-VT4) end up driving a common load, this results in a noise gain of 3 dB. As a result, the amplifier noise turned out to be approximately three times (10 dB) less than that of amplifiers whose input stage is based on the K157UD2 op-amp.

The OOS signal from the output is fed to the connection point R13R14. The gain of the CFO circuit is determined by the chains R10R13C3 and R11R1404 together with the gain control R12, which sets the gain of the device in the range of 2-5. If desired, the gain control range can be expanded by reducing R10 and R11.

Capacitors C5-C7 correct the frequency response of the amplifier in order to obtain the best transient response, but its performance is maintained without them, however, the front of a rectangular pulse in their absence acquires a slight overshoot, and ripples appear on the "shelf".

Resistors R19, R20 protect VT7, VT8 from overload in case of a short circuit at the output.

DC amplifier modes are stabilized both locally (R13, R14, R8, R9, R15, R16) and deep (about 66 dB) general OOS, due to which temperature fluctuations and drift of element parameters have little effect on its operation.

Field-effect transistors should be selected in pairs according to the initial drain current. For transistors VT1, VT2, it should be about 0.8-1.8 mA, for VT3, VT4 - at least 5-6 mA. VT1 can be taken with indices B, A, VT2 - with indices I, E, F, K, VT3, VT4 - with indices D, G, E, KT3107 - with indices B or I, KT3102 - respectively A or B, C, D, VT5-VT8 can not be selected

Capacitors C5, C7 - types KT, KD, C1-C4 - K73-16, K73-17, K71-4, K76-5, etc. As C3, C4, you can use electrolytic capacitors, for example, K50-16, K50-6 or imported.

Amplifier power supply - from any stabilized bipolar voltage source ±15 V.

Setting up an amplifier assembled from serviceable parts is easy. By selecting R8 and R9, the voltages indicated on the diagram on the drains VT1 and VT2 (12 ± 0.5 V) are set, and by selecting R17, R18 - the voltages at the emitters VT7, VT8 (0.8-1.2 V). In parallel with this, by adjusting R4, the output voltage is set close to zero.

If the desired modes of the transistors cannot be immediately set, you should first set up the input stage separately. To do this, the amplifier output is connected to a common wire (to turn off the common OOS) and the VT5 and VT6 bases are disconnected from the VT1 and VT2 drains, then shorting these bases with their emitters. After that, the modes are achieved in the input stage, as indicated above. If this succeeds, then the circuit connections are restored and R17, R18 and R4 are finally selected.

The scheme of the volume and tone control using the amplifier shown in Fig. 3 is shown in Fig. 4, where A1, A2 are two such amplifiers; PRT - physiological tone control; TKRG is a thinly compensated volume control, the output of which is connected to the UMZCH. Infrasonic frequencies are cut off in each of the amplifiers A1 and A2 both at the input (HPF R1-R3C1 and R4-R5-R6-C2, Fig. 3) and in the OOS circuit (R10-R13-C3 and R11-R14-C4) , which results in a 4th order HPF (and together with the UMZCH input HPF - a 5th order), this is enough to effectively suppress low-frequency noise with a frequency of less than 20 Hz, such as, for example, from warped records.

There is no urgent need to bypass the PSF, since it is easy to obtain a strictly horizontal frequency response by its adjustment elements. However, this function is easy to implement, as shown in Fig. 4, using switch S1 and divider R1R2.

As R12 (Fig. 3), a double variable resistor is used, the "halves" of which are included in different channels of the stereo path. In the A1 cascades, they are connected "in phase" (the resistance of the rheostat R12 in both channels changes in one direction when the regulator slider is moved) and act as an additional level regulator, thereby increasing the overload capacity of the BUZCH up to 26 dB and ensuring the matching of the frequency response of the TKRG with the signal level. In the A2 cascades, they are included "out of phase" (the resistance R12 in one channel increases, in the other it decreases) and play the role of a stereo balance regulator.

Figure 5 shows a schematic diagram of a TKRG made on a dual variable resistor with two taps of the SP3-30V type. Often in TKRG circuits, the connection of frequency correction circuits to the potentiometer engine is used. The moving pins of a motor cannot be perfect, and when the volume is adjusted, their resistance changes from almost zero to quite noticeable, especially after extended use. In a simple (not thinly compensated) regulator, this is almost not felt, especially if the subsequent stage has a sufficiently large input impedance, and may manifest itself as slight rustling during regulation.

In TKRG with the connection of correction circuits to the engine, things are much worse, the frequency response when the contact deteriorates can be distorted very strongly and become completely unacceptable, sometimes stunning the listener with a sharp sound of unnatural coloring. TKRG also suffers from frequency response distortions, the correction circuits of which are connected both to the taps and to the engine. In such TKRG, even with ideal constant contact of the engine, annoying changes in the frequency response are clearly visible by ear when the engine passes by the tap.

The proposed TKRG is devoid of these shortcomings, since in it the frequency correction circuits are not connected to the potentiometer engine. Its frequency response is shown in Fig.6. They are a good approximation to the required ones, thanks to the detailed study of the frequency-dependent links.

Electrolytic capacitors cannot be used in the TKRG circuit (and in the FRT), since the constant component of the voltage on their plates during the operation of these circuits is zero. The same types of non-electrolytic capacitors as indicated in the amplifier circuit should be used. The described preamplifier and volume and tone control unit, when working together with UMZCH, equipped with good acoustic systems, provide excellent sound.

Literature

1. Matyushkin V.P. Superlinear UMZCH class Hgh-End on transistors // Radiumator.-1998.-No. 8.-S.10-11; No. 9.-S. 10-11.

2. Matyushkin V.P. Parallel feedback loops and their application in ultrasound // Radioamator.-2000.-No. 12.-2001; №1-3.®

The schemes and designs of highly sensitive microphones in combination with self-made low-noise low-frequency amplifiers (ULF) are considered.

The design of a sensitive and low-noise amplifier (ULF) has its own characteristics. The greatest influence on the quality of sound reproduction and speech intelligibility is exerted by the amplitude-frequency characteristic (AFC) of the amplifier, its noise level, the parameters of the microphone (frequency response, directivity pattern, sensitivity, etc.) or sensors replacing it, as well as their mutual consistency with the amplifier . The amplifier must have sufficient gain.

When using a microphone, this is 60db-80db, i.e. 1000-10000 times. Taking into account the peculiarities of receiving a useful signal and its low value under conditions of a relatively significant level of interference that always exists, it is advisable in the design of the amplifier to provide for the possibility of correcting the frequency response, those. frequency selection of the processed signal.

It should be taken into account that the most informative section of the audio range is concentrated in the band from 300 Hz to 3-3.5 kHz. True, sometimes in order to reduce interference, this band is reduced even more. The use of a band-pass filter as part of an amplifier allows you to significantly increase the listening range (by 2 or more times).

Even greater range can be achieved by using selective filters with high quality factor as part of the ULF, which make it possible to isolate or suppress the signal at certain frequencies. This makes it possible to significantly improve the signal-to-noise ratio.

elemental base

Modern element base allows you to create high-quality ULF based on low-noise operational amplifiers(OU), for example, K548UN1, K548UN2, K548UNZ, KR140UD12, KR140UD20, etc.

However, despite the wide range of specialized microcircuits and op-amps, and their high parameters, ULF on transistors have not lost their significance. The use of modern, low-noise transistors, especially in the first stage, allows you to create amplifiers that are optimal in terms of parameters and complexity: low-noise, compact, economical, designed for low-voltage power supply. Therefore, transistorized ULFs often turn out to be a good alternative to amplifiers based on integrated circuits.

To minimize the noise level in amplifiers, especially in the first stages, it is advisable to use high-quality elements. These elements include low-noise bipolar transistors with high gain, for example, KT3102, KT3107. However, depending on the purpose of the ULF, field-effect transistors are also used.

The parameters of other elements are also of great importance. In low-noise cascades of electronic circuits, oxide capacitors K53-1, K53-14, K50-35, etc. are used, non-polar - KM6, MBM, etc., resistors - no worse than traditional 5% MLT-0.25 and ML T- 0.125, the best resistor option is wirewound, non-inductive resistors.

The input impedance of the ULF must correspond to the resistance of the signal source - a microphone or a sensor replacing it. Usually, they try to make the ULF input impedance equal (or slightly higher) to the resistance of the source-signal converter at the main frequencies.

To minimize electrical interference, it is advisable to use shielded wires of a minimum length to connect the microphone to the ULF. It is recommended to mount the IEC-3 electret microphone directly on the board of the first stage of the microphone amplifier.

If it is necessary to significantly remove the microphone from the ULF, an amplifier with a differential input should be used, and the connection should be made with a twisted pair of wires in the screen. The screen is connected to the circuit at one point of the common wire as close as possible to the first op-amp. This minimizes the level of electrical noise induced in the wires.

Low-noise ULF for a microphone on K548UN1A

Figure 1 shows an example of a ULF based on a specialized microcircuit - IS K548UN1A, containing 2 low-noise op-amps. The op amp and ULF, created on the basis of these op amps (IS K548UN1A), are designed for a unipolar supply voltage of 9V - ZOV. In the above ULF scheme, the first op-amp is included in the version that provides the minimum noise level of the op-amp.

Rice. 1. ULF circuit on the K548UN1A op-amp and options for connecting microphones: a - ULF on the K548UN1A op-amp, b - connecting a dynamic microphone, c - connecting an electret microphone, d - connecting a remote microphone.

Elements for the circuit in Figure 1:

• R1=240-510, R2=2.4k, R3=24k-51k (gain trim),
• R4=3k-10k, R5=1k-3k, R6=240k, R7=20k-100k (gain trim), R8=10; R9=820-1.6k (for 9V);
• C1=0.2-0.47, C2=10uF-50uF, C3=0.1, C4=4.7uF-50uF,
• C5=4.7uF-50uF, C6=10uF-50uF, C7=10uF-50uF, C8=0.1-0.47, C9=100uF-500uF;
• Op-amps 1 and 2 - op-amps IS K548UN1A (B), two op-amps in one IC package;
• T1, T2 - KT315, KT361 or KT3102, KT3107 or similar;
• T - TM-2A.

The output transistors of this ULF circuit operate without initial bias (from Irest = 0). Step-type distortions are practically absent due to the deep negative feedback covering the second op-amp of the microcircuit and output transistors. two resistors of 3-5k each from the bases of transistors to a common wire and a power wire.

By the way, in ULF in push-pull output stages without an initial bias, outdated germanium transistors work well. This makes it possible to use an op-amp with such an output stage structure with a relatively low output voltage slew rate without the risk of distortion associated with zero quiescent current. To eliminate the danger of excitation of the amplifier at high frequencies, a capacitor C3 is used, connected next to the op-amp, and the R8C8 circuit at the ULF output (quite often RC at the amplifier output can be excluded).

Low-noise microphone ULF on transistors

Figure 2 shows an example ULF circuits on transistors. In the first stages, the transistors operate in the microcurrent mode, which minimizes the internal noise of the ULF. Here it is advisable to use transistors with a large gain, but a small reverse current.

It can be, for example, 159HT1B (Ik0=20nA) or KT3102 (Ik0=50nA), or similar.

Rice. Fig. 2. ULF circuit on transistors and options for connecting microphones: a ULF on transistors, b - connection of a dynamic microphone, c - connection of an electret microphone, d - connection of a remote microphone.

Elements for the circuit in Figure 2:

• R3=5.6k-6.8k (volume control), R4=3k, R5=750,
• R6=150k, R7=150k, R8=33k; R9=820-1.2k, R10=200-330,
• R11=100k (adjustment, Uet5=Uet6=1.5V),
• R12 \u003d 1 k (adjustment of the quiescent current T5 and T6, 1-2 mA);
• C1=10uF-50uF, C2=0.15uF-1uF, C3=1800,
• C4=10uF-20uF, C5=1uF, C6=10uF-50uF, C7=100uF-500uF;
• T1, T2, T3 -159NT1 V, KT3102E or similar,
• T4, T5 - KT315 or similar, but MP38A is also possible,
• T6 - KT361 or similar, but MP42B is also possible;
• M - MD64, MD200 (b), IEC-3 or similar (c),
• T - TM-2A.

The use of such transistors makes it possible to ensure not only stable operation of transistors at low collector currents, but also to achieve good amplifying characteristics at a low noise level.

Output transistors can be used both silicon (KT315 and KT361, KT3102 and KT3107, etc.) and germanium (MP38A and MP42B, etc.). Setting up the circuit is reduced to setting resistor R2 and resistor R3 of the corresponding voltages on the transistors: 1.5V - on the collector T2 and 1.5V - on the emitters T5 and T6.

Op-amp microphone amplifier with differential input

Figure 3 shows an example of ULF on Op-amp with differential input. A properly assembled and tuned ULF provides significant suppression of common mode noise (60 dB or more). This ensures the selection of a useful signal with a significant level of common mode noise.

It should be recalled that common-mode interference is interference that arrives in equal phases at both inputs of the ULF op-amp, for example, interference induced on both signal wires from a microphone. To ensure the correct operation of the differential stage, it is necessary to exactly fulfill the condition: R1 = R2, R3 = R4.

Fig.3. ULF circuit on an op-amp with a differential input and options for connecting microphones: a - ULF with a differential input, b - connecting a dynamic microphone, c - connecting an electret microphone, d - connecting a remote microphone.

Elements for the circuit in Figure 3:

• R7=47k-300k (gain adjustment, K=1+R7/R6), R8=10, R9=1.2k-2.4k;
• C1=0.1-0.22, C2=0.1-0.22, SZ=4.7uF-20uF, C4=0.1;
• OU - KR1407UD2, KR140UD20, KR1401UD2B, K140UD8 or other OU in a typical inclusion, preferably with internal correction;
• D1 - zener diode, for example, KS133, you can use the LED in a normal turn on, for example, AL307;
• M - MD64, MD200 (b), IEC-3 or similar (c),
• T - TM-2A.

It is advisable to select resistors using an ohmmeter among 1% resistors with good temperature stability. To ensure the necessary balance, it is recommended that one of the four resistors (for example, R2 or R4) be made variable. It can be a high-precision variable resistor trimmer with an internal gear.

To minimize noise, the input impedance of the VLF (resistors R1 and R2) must correspond to the resistance of the microphone or the sensor replacing it. ULF output transistors operate without initial bias (from 1 rest = 0). Step-type distortion is practically absent due to deep negative feedback covering the second op-amp and output transistors. If necessary, the switching circuit of transistors can be changed.

Setting up the differential stage: apply a sinusoidal signal of 50 Hz to both inputs of the differential channel at the same time, by selecting the value of R3 or R4, ensure a zero signal level of 50 Hz at the output of op-amp 1. A 50 Hz signal is used for tuning, since the 50 Hz mains gives the maximum contribution to the total interference voltage. Good resistors and careful tuning can achieve common mode rejection of 60dB-80dB or more.

To increase the stability of the ULF, it is advisable to shunt the power supply terminals of the op-amp with capacitors and turn on the RC-element at the amplifier output (as in the amplifier circuit in Figure 1). For this purpose, you can use capacitors KM6.

A twisted pair of wires in the screen was used to connect the microphone. The screen is connected to the ULF (only at one point !!) as close as possible to the input of the op-amp.

Improved amplifiers for sensitive microphones

The use of low-speed op amps in the ULF output stages and the operation of silicon transistors in power amplifiers in the mode without initial bias (the quiescent current is zero - mode B) can, as noted above, lead to transient distortions of the "step" type. In this case, to eliminate these distortions, it is advisable to change the structure of the output stage so that the output transistors operate with a small initial current (AB mode).

Figure 4 shows an example of such an upgrade of the above differential input amplifier circuit (Figure 3).

Fig.4. ULF circuit on an op-amp with a differential input and a low-distortion output stage.

Elements for the circuit in Figure 4:

• R1=R2=20k (equal to or slightly higher than the maximum source impedance in the operating frequency range),
• RЗ=R4=1m-2m; R5=2k-10k, R6=1k-Zk,
• R7=47k-300k (gain adjustment, K=1+R7/R6),
• R8=10, R10=10k-20k, R11=10k-20k;
• C1=0.1-0.22, C2=0.1-0.22, C3=4.7uF-20uF, C4=0.1;
• OU - K140UD8, KR1407UD2, KR140UD12, KR140UD20, KR1401UD2B or other OU in a typical inclusion and preferably with internal correction;
• T1, T2 - KT3102, KT3107 or KT315, KT361, or similar;
• D2, D3 - KD523 or similar;
• M - MD64, MD200, IEC-3 or similar (c),
• T - TM-2A.

Figure 5 shows an example ULF on transistors. In the first stages, the transistors operate in the microcurrent mode, which minimizes ULF noise. The circuit is in many respects similar to the circuit in Figure 2. To increase the share of a useful low-level signal against the background of inevitable interference, a band-pass filter is included in the ULF circuit, which ensures the selection of frequencies in the 300 Hz -3.5 kHz band.

Fig.5. ULF circuit on transistors with a band-pass filter and options for connecting microphones: a - ULF with a band-pass filter, b - connection of a dynamic microphone, c - connection of an electret microphone.

Elements for the circuit in Figure 5:

• R1=43k-51k, R2=510k (adjustment, Ukt2=1.2V-1.8V),
• R3=5.6k-6.8k (volume control), R4=3k, R5=8.2k,
• R6=8.2k, R7=180, R8=750; R9=150k, R10=150k, R11=33k,
• R12=620, R13=820-1.2k, R14=200-330,
• R15=100k (adjustment, Uet5=Uet6=1.5V), R16=1k (adjustment of quiescent current T5 and T6, 1-2mA);
• C1=10uF-50uF, C2=0.15-0.33, C3=1800,
• C4=10uF-20uF, C5=0.022, C6=0.022,
• C7=0.022, C8=1uF, C9=10uF-20uF, C10=100uF-500uF;
• T1, T2, T3 -159NT1 V, KT3102E or similar;
• T4, T5 - KT3102, KT315 or similar, but outdated, germanium transistors, for example, MP38A,
• T6 - KT3107 (if T5 - KT3102), KT361 (if T5 - KT315) or similar, but outdated, germanium transistors, for example, MP42B (if T5 - MP38A);
• M - MD64, MD200 (b), IEC-3 or similar (c),
• T - TM-2A.

In this circuit, it is also advisable to use transistors with a high gain, but a small reverse collector current (Ik0), for example, 159NT1V (Ik0 \u003d 20nA) or KT3102 (Ik0 \u003d 50nA), or similar. Output transistors can be used both silicon (KT315 and KT361, KT3102 and KT3107, etc.) and germanium (obsolete transistors MP38A and MP42B, etc.).

Setting up the circuit, as in the case of the ULF circuit in Fig. 11.2, comes down to setting resistor R2 and resistor R3 of the corresponding voltages on transistors T2 and T5, T6: 1.5V - on the collector T2 and 1.5V - on the emitters T5 and T6.

Microphone design

From a large sheet of thick paper with pile, under velvet, a pipe is made with a diameter of 10-15 cm and a length of 1.5-2 m. The pile, as you might guess, of course, should not be outside, but inside. A sensitive microphone is inserted into one end of this tube. It is better if it is a good dynamic or condenser microphone.

However, you can use a regular, household, microphone. It can be, for example, a dynamic microphone such as MD64, MD200 or even a miniature MKE-3.

True, with a household microphone, the result will be somewhat worse. Of course, the microphone must be connected with a shielded cable to a sensitive amplifier with a low level of self-noise (Fig. 1 and 2). If the cable length exceeds 0.5 m, then it is better to use a microphone amplifier that has a differential input, for example, a ULF on an op amp (Fig.

This will reduce the common-mode component of interference - various kinds of interference from nearby electromagnetic devices, a 50 Hz background from a 220 V network, etc. Now about the second end of this paper pipe. If this free end of the pipe is directed to a sound source, for example, to a group of people talking, then speech can be heard. It would seem nothing special.

That's what microphones are for. And you don't even need a pipe for that. However, it is surprising that the distance to the speakers can be significant, for example, 100 meters or more. Both the amplifier and the microphone, equipped with such a pipe, allow everything to be heard quite well at such a considerable distance.

The distance can even be increased by using special selective filters that make it possible to isolate or suppress the signal in narrow frequency bands.

This makes it possible to increase the level of the useful signal in the face of inevitable interference. In a simplified version, instead of special filters, you can use a bandpass filter in the ULF (Fig. 4) or use a conventional equalizer - a multi-band tone control, in extreme cases - a traditional one, t.s. conventional, two-way, bass and treble tone control.

Literature: Rudomedov E.A., Rudometov V.E. - Electronics and espionage passions-3.