Self-excitation of the output stage of a transistor amplifier. Transistor amplifier: types, circuits, simple and complex

There were already publications on Habré about DIY tube amplifiers, which were very interesting to read. No doubt, they sound wonderful, but for everyday use it is easier to use a transistor device. Transistors are more convenient because they do not require warming up before operation and are more durable. And not everyone dares to start a lamp saga with anode potentials under 400 V, and transistor transformers for a couple of tens of volts are much safer and simply more affordable.

I chose John Linsley Hood's 1969 circuit as the circuit to reproduce, taking the author's parameters based on the impedance of my speakers 8 ohms.

The classic circuit from a British engineer, published almost 50 years ago, is still one of the most reproducible and collects exceptionally positive reviews about itself. There are many explanations for this:
- the minimum number of elements simplifies installation. It is also believed that the simpler the design, the better the sound;
- despite the fact that there are two output transistors, they do not need to be sorted into complementary pairs;
- output of 10 watts with a margin is enough for ordinary human dwellings, and an input sensitivity of 0.5-1 volts is very well consistent with the output of most sound cards or players;
- class A - it is also class A in Africa, if we are talking about good sound. About comparison with other classes will be a little lower.

Internal design

The amplifier starts with power. Separation of two channels for stereo is best done from two different transformers, but I limited myself to one transformer with two secondary windings. After these windings, each channel exists on its own, so we must not forget to multiply everything mentioned below by two. On the breadboard we make bridges on Schottky diodes for the rectifier.

It is possible on ordinary diodes or even ready-made bridges, but then they need to be shunted with capacitors, and the voltage drop across them is greater. After the bridges, there are CRC filters of two 33,000 microfarad capacitors and a 0.75 ohm resistor between them. If you take both the capacitance and the resistor less, then the CRC filter will become cheaper and heat up less, but the ripple will increase, which is not comme il faut. These parameters, IMHO, are reasonable in terms of price-effect. A powerful cement resistor is needed in the filter, with a quiescent current of up to 2A it will dissipate 3 W of heat, so it is better to take it with a margin of 5-10 W. For the rest of the resistors in the power circuit, 2 W will be enough.

Next, we move on to the amplifier board itself. A lot of ready-made kits are sold in online stores, but there are no less complaints about the quality of Chinese components or illiterate layouts on the boards. Therefore, it is better to do it yourself, under your own “loose”. I made both channels on a single breadboard, so that later I can attach it to the bottom of the case. Run with test items:

Everything except the output transistors Tr1/Tr2 is located on the board itself. Output transistors are mounted on radiators, more on that below. To the author's scheme from the original article, you need to make the following remarks:

Not everything needs to be soldered right away. It is better to put resistors R1, R2 and R6 first with trimmers, after all the adjustments, remove them, measure their resistance and solder the final fixed resistors with the same resistance. The setting is reduced to the following operations. First, using R6, it is set so that the voltage between X and zero is exactly half of the voltage + V and zero. In one of the channels, I lacked 100 kOhm, so it's better to take these trimmers with a margin. Then, with the help of R1 and R2 (keeping their approximate ratio!) the quiescent current is set - we put the tester to measure direct current and measure this very current at the input point of the plus supply. I had to significantly reduce the resistance of both resistors to get the desired quiescent current. The quiescent current of the amplifier in class A is maximum and, in fact, in the absence of an input signal, everything goes into thermal energy. For 8 ohm speakers, this current, according to the author's recommendation, should be 1.2 A at 27 volts, which means 32.4 watts of heat per channel. Since it can take several minutes for the current to be applied, the output transistors must already be on cooling heatsinks or they will quickly overheat and die. Because they get hot most of the time.

It is possible that, as an experiment, you will want to compare the sound of different transistors, so you can also leave the possibility of a convenient replacement for them. I tried on the input 2N3906, KT361 and BC557C, there was a slight difference in favor of the latter. In the pre-weekend, we tried KT630, BD139 and KT801, settled on imported ones. Although all of the above transistors are very good, and the difference can be rather subjective. At the output, I immediately put 2N3055 (ST Microelectronics), since many people like them.

When adjusting and lowering the resistance of the amplifier, the cutoff frequency of the low frequencies may increase, so for the capacitor at the input it is better to use not 0.5 microfarads, but 1 or even 2 microfarads in a polymer film. The Russian picture-scheme “Ultralinear Class A Amplifier” is still circulating on the Web, where this capacitor is generally proposed as 0.1 microfarads, which is fraught with a cutoff of all basses at 90 Hz:

They write that this circuit is not prone to self-excitation, but just in case, a Zobel circuit is placed between the X point and the ground: R 10 Ohm + C 0.1 microfarad.
- fuses, they can and should be installed both on the transformer and on the power input of the circuit.
- it would be very appropriate to use thermal paste for maximum contact between the transistor and the heatsink.

Locksmith and carpentry

Now about the traditionally most difficult part in DIY - the case. The dimensions of the case are set by radiators, and in class A they should be large, remember about 30 watts of heat on each side. At first, I underestimated this power and made a case with average radiators 800cm² per channel. However, with a set quiescent current of 1.2A, they heated up to 100 ° C in just 5 minutes, and it became clear that something more powerful was needed. That is, you need to either install larger radiators, or use coolers. I didn’t want to make a quadcopter, so I bought giant handsome HS 135-250s with an area of ​​2500 cm² for each transistor. As practice has shown, such a measure turned out to be a little redundant, but now the amplifier can be safely touched by hands - the temperature is only 40 ° C even in rest mode. Drilling holes in the radiators for fasteners and transistors became a problem - the originally purchased Chinese metal drills were drilled extremely slowly, it would take at least half an hour for each hole. Cobalt drills with a sharpening angle of 135 ° from a well-known German manufacturer came to the rescue - each hole is passed in a few seconds!

I made the body out of Plexiglas. We immediately order cut rectangles from glaziers, make the necessary holes for fastenings in them and paint the reverse side with black paint.

The plexiglass painted on the back looks very nice. Now it remains only to assemble everything and enjoy the music ... oh yes, during the final assembly, it is also important to properly dilute the ground to minimize the background. As it was found out decades before us, C3 needs to be connected to the signal ground, i.e. to the minus of the input-input, and all other minuses can be sent to the "star" near the filter capacitors. If everything is done correctly, then no background can be heard, even if you bring your ear to the speaker at maximum volume. Another "ground" feature that is typical for sound cards that are not galvanically isolated from the computer is interference from the motherboard, which can creep through USB and RCA. Judging by the Internet, the problem is common: in the speakers you can hear the sounds of the HDD, printer, mouse and the background of the system unit's power supply. In this case, the easiest way is to break the ground loop by taping the ground on the amplifier plug with electrical tape. There is nothing to fear here, because. there will be a second ground loop through the computer.

I didn’t make a volume control on the amplifier, because I couldn’t get any high-quality ALPS, and I didn’t like the rustling of Chinese potentiometers. Instead, a conventional 47 kΩ resistor was installed between the “ground” and the “signal” of the input. Moreover, the regulator of an external sound card is always at hand, and each program also has a slider. Only the vinyl player does not have a volume control, so to listen to it, I attached an external potentiometer to the connecting cable.

I can guess this container in 5 seconds...

Finally, you can start listening. The sound source is Foobar2000 → ASIO → external Asus Xonar U7. Speakers Microlab Pro3. The main advantage of these speakers is a separate block of their own amplifier on the LM4766 chip, which can be immediately removed somewhere far away. Much more interesting with this acoustics sounded the amplifier from the Panasonic mini-system with the proud inscription Hi-Fi or the amplifier of the Soviet player Vega-109. Both of the above devices work in class AB. The JLH presented in the article outplayed all of the above comrades in one wicket, according to the results of a blind test for 3 people. Although the difference was audible to the naked ear and without any tests, the sound is clearly more detailed and transparent. It's quite easy, for example, to hear the difference between 256kbps MP3 and FLAC. I used to think that the lossless effect is more like a placebo, but now the opinion has changed. Similarly, it became much more pleasant to listen to files not compressed from loudness war - dynamic range less than 5 dB is not ice at all. The Linsley Hood is worth the time and money, as a similar branded amp will cost a lot more.

Material costs

Transformer 2200 rub.
Output transistors (6 pieces with a margin) 900 rubles.
Filter capacitors (4 pcs) 2700 r.
"Rose" (resistors, small capacitors and transistors, diodes) ~ 2000 rubles.
Radiators 1800 r.
Plexiglas 650 rub.
Paint 250 rub.
Connectors 600 rub.
Boards, wires, silver solder, etc. ~1000 r.
TOTAL ~12100 rub.

The simplest transistor amplifier can be a good tool for studying the properties of devices. The schemes and designs are quite simple, you can independently manufacture the device and check its operation, measure all parameters. Thanks to modern field-effect transistors, it is possible to make a miniature microphone amplifier literally from three elements. And connect it to a personal computer to improve the sound recording parameters. And the interlocutors during conversations will hear your speech much better and more clearly.

Frequency characteristics

Low-frequency (sound) frequency amplifiers are available in almost all household appliances - music centers, televisions, radios, radios, and even personal computers. But there are also high-frequency amplifiers on transistors, lamps and microcircuits. Their difference is that ULF allows you to amplify the signal of only the audio frequency, which is perceived by the human ear. Transistor audio amplifiers allow you to reproduce signals with frequencies in the range from 20 Hz to 20,000 Hz.

Therefore, even the simplest device is able to amplify the signal in this range. And it does it as evenly as possible. The gain depends directly on the frequency of the input signal. The graph of the dependence of these quantities is almost a straight line. If, on the other hand, a signal with a frequency outside the range is applied to the input of the amplifier, the quality of work and the efficiency of the device will quickly decrease. ULF cascades are assembled, as a rule, on transistors operating in the low and medium frequency ranges.

Classes of operation of audio amplifiers

All amplifying devices are divided into several classes, depending on what degree of current flow through the cascade during the period of operation:

1. Class "A" - the current flows non-stop during the entire period of operation of the amplifying stage.
2. In the class of work "B" current flows for half the period.
3. Class "AB" indicates that the current flows through the amplifying stage for a time equal to 50-100% of the period.
4. In "C" mode, the electric current flows for less than half of the operating time.
5. Mode "D" ULF has been used in amateur radio practice quite recently - a little over 50 years. In most cases, these devices are implemented on the basis of digital elements and have a very high efficiency - over 90%.

The presence of distortion in various classes of low-frequency amplifiers

The working area of ​​a class "A" transistor amplifier is characterized by rather small non-linear distortions. If the incoming signal throws out higher voltage pulses, this causes the transistors to saturate. In the output signal, higher harmonics (up to 10 or 11) begin to appear near each harmonic. Because of this, a metallic sound, characteristic only for transistor amplifiers, appears.

With an unstable power supply, the output signal will be modeled in amplitude near the mains frequency. The sound will become harsher on the left side of the frequency response. But the better the power stabilization of the amplifier, the more complex the design of the entire device becomes. ULF operating in class "A" have a relatively low efficiency - less than 20%. The reason is that the transistor is constantly on and current flows through it constantly.

To increase (albeit insignificant) efficiency, you can use push-pull circuits. One disadvantage is that the half-waves of the output signal become asymmetrical. If you transfer from class "A" to "AB", the non-linear distortion will increase by 3-4 times. But the efficiency of the entire circuit of the device will still increase. ULF classes "AB" and "B" characterizes the increase in distortion with a decrease in the signal level at the input. But even if you turn up the volume, it will not help to completely get rid of the shortcomings.

Work in intermediate classes

Each class has several varieties. For example, there is a class of amplifiers "A +". In it, the transistors at the input (low-voltage) operate in the "A" mode. But high-voltage, installed in the output stages, work either in "B" or in "AB". Such amplifiers are much more economical than those operating in class "A". A noticeably smaller number of non-linear distortions - no higher than 0.003%. Better results can be achieved using bipolar transistors. The principle of operation of amplifiers on these elements will be discussed below.

But still there are a large number of higher harmonics in the output signal, which makes the sound characteristic metallic. There are also amplifier circuits that work in the "AA" class. In them, non-linear distortion is even less - up to 0.0005%. But the main drawback of transistor amplifiers is still there - a characteristic metallic sound.

"Alternative" designs

It cannot be said that they are alternative, just some specialists involved in the design and assembly of amplifiers for high-quality sound reproduction are increasingly preferring tube designs. Tube amplifiers have the following advantages:

1. Very low level of non-linear distortion in the output signal.
2. There are fewer higher harmonics than in transistor designs.

But there is one huge minus that outweighs all the advantages - you must definitely install a device for coordination. The fact is that the tube cascade has a very high resistance - several thousand ohms. But the speaker winding resistance is 8 or 4 ohms. To match them, you need to install a transformer.

Of course, this is not a very big drawback - there are also transistor devices that use transformers to match the output stage and the speaker system. Some experts argue that the most effective circuit is hybrid - which uses single-ended amplifiers that are not covered by negative feedback. Moreover, all these cascades operate in the ULF class "A" mode. In other words, a transistorized power amplifier is used as a repeater.

Moreover, the efficiency of such devices is quite high - about 50%. But you should not focus only on efficiency and power indicators - they do not speak of the high quality of sound reproduction by the amplifier. Much more important are the linearity of the characteristics and their quality. Therefore, you need to pay attention first of all to them, and not to power.

Scheme of a single-ended ULF on a transistor

The simplest amplifier, built according to the common emitter circuit, operates in class "A". The circuit uses a semiconductor element with an n-p-n structure. A resistance R3 is installed in the collector circuit, which limits the flowing current. The collector circuit is connected to the positive power wire, and the emitter circuit is connected to the negative. In the case of using semiconductor transistors with a p-n-p structure, the circuit will be exactly the same, only the polarity will need to be reversed.

With the help of a coupling capacitor C1, it is possible to separate the AC input signal from the DC source. In this case, the capacitor is not an obstacle to the flow of alternating current along the base-emitter path. The internal resistance of the emitter-base junction, together with resistors R1 and R2, is the simplest supply voltage divider. Typically, resistor R2 has a resistance of 1-1.5 kOhm - the most typical values ​​\u200b\u200bfor such circuits. In this case, the supply voltage is divided exactly in half. And if you power the circuit with a voltage of 20 Volts, you can see that the value of the current gain h21 will be 150. It should be noted that HF ​​amplifiers on transistors are made according to similar circuits, only they work a little differently.

In this case, the emitter voltage is 9 V and the drop in the “E-B” circuit section is 0.7 V (which is typical for transistors based on silicon crystals). If we consider an amplifier based on germanium transistors, then in this case the voltage drop in the “E-B” section will be equal to 0.3 V. The current in the collector circuit will be equal to that which flows in the emitter. You can calculate by dividing the emitter voltage by the resistance R2 - 9V / 1 kOhm = 9 mA. To calculate the value of the base current, it is necessary to divide 9 mA by the gain h21 - 9mA / 150 \u003d 60 μA. ULF designs usually use bipolar transistors. The principle of its work is different from the field.

On the resistor R1, you can now calculate the drop value - this is the difference between the base and supply voltages. In this case, the base voltage can be found by the formula - the sum of the characteristics of the emitter and the "E-B" transition. When powered by a 20 Volt source: 20 - 9.7 \u003d 10.3. From here, you can calculate the resistance value R1 = 10.3V / 60 μA = 172 kOhm. The circuit contains capacitance C2, which is necessary for the implementation of the circuit through which the alternating component of the emitter current can pass.

If you do not install capacitor C2, the variable component will be very limited. Because of this, such a transistor audio amplifier will have a very low current gain h21. It is necessary to pay attention to the fact that in the above calculations the base and collector currents were assumed to be equal. Moreover, the base current was taken to be the one that flows into the circuit from the emitter. It occurs only when a bias voltage is applied to the output of the base of the transistor.

But it must be borne in mind that absolutely always, regardless of the presence of bias, the collector leakage current necessarily flows through the base circuit. In circuits with a common emitter, the leakage current is increased by at least 150 times. But usually this value is taken into account only when calculating amplifiers based on germanium transistors. In the case of using silicon, in which the current of the "K-B" circuit is very small, this value is simply neglected.

MIS transistor amplifiers

The field-effect transistor amplifier shown in the diagram has many analogues. Including using bipolar transistors. Therefore, we can consider as a similar example the design of a sound amplifier assembled according to a common emitter circuit. The photo shows a circuit made according to a circuit with a common source. R-C connections are assembled on the input and output circuits so that the device operates in the class “A” amplifier mode.

Alternating current from the signal source is separated from the DC supply voltage by capacitor C1. Be sure the field-effect transistor amplifier must have a gate potential that will be lower than that of the source. In the presented diagram, the gate is connected to a common wire through a resistor R1. Its resistance is very large - resistors of 100-1000 kOhm are usually used in designs. Such a large resistance is chosen so that the signal at the input is not shunted.

This resistance almost does not pass electric current, as a result of which the potential of the gate (in the absence of a signal at the input) is the same as that of the ground. At the source, the potential is higher than that of the ground, only due to the voltage drop across the resistance R2. From this it is clear that the potential of the gate is lower than that of the source. Namely, this is required for the normal functioning of the transistor. It should be noted that C2 and R3 in this amplifier circuit have the same purpose as in the design discussed above. And the input signal is shifted relative to the output signal by 180 degrees.

ULF with output transformer

You can make such an amplifier with your own hands for home use. It is carried out according to the scheme that works in class "A". The design is the same as discussed above - with a common emitter. One feature - it is necessary to use a transformer for matching. This is a disadvantage of such a transistor audio amplifier.

The collector circuit of the transistor is loaded with a primary winding, which develops an output signal transmitted through the secondary to the speakers. A voltage divider is assembled on resistors R1 and R3, which allows you to select the operating point of the transistor. With the help of this circuit, a bias voltage is supplied to the base. All other components have the same purpose as the circuits discussed above.

push-pull audio amplifier

This is not to say that this is a simple transistor amplifier, since its operation is a little more complicated than that of those discussed earlier. In push-pull ULF, the input signal is split into two half-waves, different in phase. And each of these half-waves is amplified by its own cascade, made on a transistor. After each half-wave has been amplified, both signals are combined and fed to the speakers. Such complex transformations can cause signal distortion, since the dynamic and frequency properties of two, even of the same type, transistors will be different.

As a result, the sound quality at the output of the amplifier is significantly reduced. When a push-pull amplifier in class "A" is working, it is not possible to reproduce a complex signal with high quality. The reason is that the increased current flows constantly through the arms of the amplifier, the half-waves are asymmetrical, and phase distortions occur. The sound becomes less intelligible, and when heated, signal distortion increases even more, especially at low and ultra-low frequencies.

Transformerless ULF

The low-frequency amplifier on a transistor, made using a transformer, despite the fact that the design may have small dimensions, is still imperfect. Transformers are still heavy and bulky, so it's best to get rid of them. A much more efficient circuit is made on complementary semiconductor elements with different types of conductivity. Most of the modern ULFs are performed exactly according to such schemes and work in class "B".

Two powerful transistors used in the design work according to the emitter follower circuit (common collector). In this case, the input voltage is transmitted to the output without loss and amplification. If there is no signal at the input, then the transistors are on the verge of turning on, but still turned off. When a harmonic signal is applied to the input, the first transistor opens with a positive half-wave, and the second one is in the cutoff mode at this time.

Therefore, only positive half-waves can pass through the load. But negative ones open the second transistor and completely block the first one. In this case, only negative half-waves are in the load. As a result, the signal amplified in power is at the output of the device. Such a transistor amplifier circuit is quite effective and is able to provide stable operation, high-quality sound reproduction.

ULF circuit on one transistor

Having studied all the above features, you can assemble an amplifier with your own hands on a simple element base. The transistor can be used domestically KT315 or any of its foreign analogues - for example BC107. As a load, you need to use headphones, the resistance of which is 2000-3000 ohms. A bias voltage must be applied to the base of the transistor through a 1 MΩ resistor and a 10 µF decoupling capacitor. The circuit can be powered from a source with a voltage of 4.5-9 Volts, current - 0.3-0.5 A.

If the resistance R1 is not connected, then there will be no current in the base and collector. But when connected, the voltage reaches a level of 0.7 V and allows a current of about 4 μA to flow. In this case, the current gain will be about 250. From here, you can make a simple calculation of the transistor amplifier and find out the collector current - it turns out to be 1 mA. Having assembled this transistor amplifier circuit, you can test it. Connect the load - headphones to the output.

Touch the input of the amplifier with your finger - a characteristic noise should appear. If it is not there, then most likely the design is assembled incorrectly. Recheck all connections and element ratings. To make the demonstration clearer, connect a sound source to the ULF input - the output from the player or phone. Listen to music and appreciate the sound quality.

class A amplifier.

Operates in linear mode: both transistors operate in the same modes. This providesminimum distortion , but as a result, low efficiency (15-30%), i.e. this class is uneconomical in terms of energy consumption and heating. The power consumption is independent of the output power.

Class B Amplifier

This class mainly includes amplifiers with output transistors of the same conductivity. Each of the transistors operates in a key mode, i.e. amplifies only its signal half-wave in a linear mode (for example, positive if transistors with N-P-N conductivity are used). In order to amplify the negative half-wave of the signal, a phase inverter is used on another transistor. It's like two separate A classes (one for each half-wave). An amplifier of this class has a high efficiency (about 70%). The power consumption of the amplifier is proportional to the output power, in the absence of a signal at the input it is equal to zero. Amplifiers of this class are rare among modern amplifiers.

Class AB Amplifier

The most common type of amplifiers. This class combines the qualities of class A and class B amplifiers, i.e. high class B efficiency and low class A non-linear distortion. the operating point is selected at the beginning of the linear section of the current-voltage characteristic. Due to this, in the absence of a signal at the inputthe amplifying elements are not locked, and some current flows through them (the so-called "quiescent current") , sometimes significant. And here there is a need to regulate and stabilize this current so that the transistors operate in the same modes without overloading each other. Incorrect setting of the quiescent current will lead to overheating of the transistors and their failure.

So: for the output stage, there are two very important parameters (and especially for class AB):

quiescent current and quiescent voltage

If transistors had an ideal characteristic (which actually does not happen), then the quiescent current could be considered equal to zero. In reality, the collector current can increase both due to the spread in the characteristics of transistors and from their temperature. Moreover: an increase in temperature can lead to an avalanche-like overheating and thermal breakdown of the transistor. The fact is that with an increase in temperature, the collector current only increases, and therefore the heating of the transistor also increases.

rest voltage: constant voltage at the junction point of the transistors (output to the load). It must be equal to "0" for a bipolar supply of the output stage, or half the supply voltage for a unipolar supply. In other words: both transistors of the output stage must have the same base bias, that is, they are open evenly, compensating each other.

These two parameters must be stabilized and, first of all, their temperature dependence must be excluded.

For this purpose, the amplifiers use an additional transistor, which is ballasted into the base circuits of the output transistors (moreover, most often it is placed directly on the radiator next to the output transistors, thereby controlling their temperature).

What is an output transistor? Output, or terminal, transistors are called transistors that are part of the design of the output (last) stages in cascade amplifiers (having at least two or three stages) of frequency. In addition to weekends, there are also preliminary cascades, that's all, some are located before the weekend.

A cascade is a transistor equipped with a resistor, capacitor and other elements that ensure its operation as an amplifier. All the number of preliminary stages available in the amplifier must provide an increase in the frequency voltage in such a way that the resulting value is suitable for the operation of the output transistor. In turn, himself output transistor increases the power of frequency oscillations to a value that ensures the operation of the dynamic head.

When assembling the most simple transistor amplifiers, the output transistor is taken as low-power as in the preliminary stages. Many find this very appropriate in terms of the ergonomics of the device. The output power readings of such an amplifier are small: from 10-20 mW to one and a half hundred.

In situations where the problem of saving is not so acute, then a transistor with higher power readings is used in the design of the output stage.

The quality of the amplifier is determined by several parameters, but the most accurate representation can be obtained from: data on the output power (P out), sensitivity and frequency response.

Measure the quiescent current of the output transistor

The quiescent current is called the collector current, which passes through the transistors of the output stages, provided that there is no signal. In conditionally ideal (impossible in fact) conditions, the value of such a current should be at zero. In fact, this is not entirely true, its own temperature and characteristic differences in different types of transistors affect this indicator. In the worst case, overheating is possible, which will cause thermal breakdown of the transistor.

In addition, there is another indicator - the voltage of rest. It shows the voltage value of the connecting point of the transistors. If the power supply of the cascade is bipolar, then the voltage will be zero, and if it is unipolar, then the voltage is 1/2 of the supply voltage.

Both of these indicators must be stabilized, and for this, temperature control should be taken care of as a priority.

An additional transistor is usually taken as a stabilizer, which is connected to the base circuits as a ballast (most often it ends up right on the radiator, as close as possible to the output transistors).

In order to discover what quiescent current of output transistors or cascades, it is necessary to measure the voltage drop data for its emitter resistors with a multimeter (values ​​are usually expressed in millivolts), and then, based on Ohm's law and real resistance data, it will be possible to calculate the desired indicator: divide the voltage drop value by the real value resistance - the value of the quiescent current for a given output transistor.

All measurements must be made very carefully, otherwise you will have to replace the transistor.

There is another way, much less traumatic. Instead of fuses, you will need to set a resistance of 100 ohms and a minimum power of 0.5 watts for each channel. In the absence of fuses, the resistance is connected to the power break. After the power is supplied to the amplifier, the readings are measured by the voltage drop at the above resistance level. Further mathematics is extremely simple: a voltage drop of 1 V corresponds to a quiescent current of 10mA. Similarly, at 3.5 V, you get 35 mA, and so on.

Classification of output stages

There are several methods for assembling the output stage:

• From transistors having different conductivity. For these purposes, "complementary" (similar in parameters) transistors are most often used.
• Of transistors having the same conductivity.
• From transistors of a composite type.
• From field effect transistors.

The operation of an amplifier designed using complementary transistors is simple: a positive signal half-wave drives one transistor, and a negative half-wave drives another. It is necessary that the shoulders (transistors) work in the same modes, and to implement this, the base bias is used.

If the amplifier uses the same transistors in operation, then this has no fundamental differences from the first option. Except for the fact that for such transistors the signal should not be different.

When working with other types of amplifiers, it must be remembered that the negative voltage is for p-n-p transistors, and the positive voltage is for n-p-n transistors.

Usually the name of the power amplifier belongs to the final stage, since it works with the largest values, although from a technical point of view, the preliminary stages can also be called that. The main indicators of the amplifier include: useful power delivered to the load, efficiency, amplified frequency band, non-linear distortion coefficient. These figures are strongly influenced output characteristic of the transistor. When creating a voltage amplifier, single-cycle and push-pull circuits can be used. In the first case, the operating mode of the amplifier is linear (class A). This situation is characterized by the fact that the current flow through the transistor lasts until the period of the input signal ends.

The single-ended amplifier is characterized by high linearity. However, these qualities can be distorted when the core is magnetized. To prevent this situation, care must be taken to have a transformer circuit with a high level of inductance for the primary circuit. This will affect the dimensions of the transformer. In addition, due to the principle of its operation, it has a rather low efficiency.

In comparison, the data for a push-pull amplifier (class B) is much higher. This mode allows you to distort the shape of the transistor current at the output. This increases the result of the ratio of alternating and direct currents, at the same time reducing the level of power consumption, this is considered the main advantage of using push-pull amplifiers. Their work is ensured by the supply of two equal in value, but phase-opposite voltages. If there is no mid-point transformer, then you can use a phase-inverted stage, which will remove the voltages opposite in phase from the corresponding resistors of the collector and emitter circuits.

There is a push-pull circuit that does not include an output transformer. This will require different types of transistors operating as emitter followers. If you act on a bipolar input signal, then the transistors will open in turn, and the currents will diverge in opposite directions.

Transistor replacement

As ULF (low frequency amplifiers) become more and more popular, it would be absolutely useful to know what to do if such a device fails.

If the output transistor heats up, then there is a high probability that it is broken or burned out. In such a situation it is necessary:

• Make sure the integrity of all other diodes and transistors included in the amplifier;
• When repairs are being carried out, it is very desirable to connect the amplifier to the network through a 40-100 V light bulb, this will help save the remaining transistors intact under any circumstances;
• First of all, the emitter-base section and transistors are bridged, then the primary diagnostics of the ULF is carried out (any changes and reactions are easily recorded using the glow of the lamp);
• The main indicator of the operating state and adequate tuning of the transistor can be considered voltage data for the base-emitter section.
• Revealing voltage data between the case and individual sections of the circuit is practically useless, it does not give any information about a possible breakdown.

Even the most simplified version of the check (before and after replacement of output transistors was produced) must necessarily include several points:

• Apply a minimum voltage to the base and emitter of the output transistor to establish a quiescent current;
• Check the effectiveness of your actions by sound or using an oscilloscope (“step” and signal distortion at a power minimum should be absent);
• Using an oscilloscope, determine the symmetry of the restrictions on the resistors at the maximum power of the amplifier.
• Make sure that the "passport" and the actual power of the amplifier match.
• It is imperative to check the working condition of the current-limiting circuits, if any, on the final stage. Here you can not do without an adjustable load resistor.

First start-up after repair work has been carried out:

1. It is undesirable to immediately install output transistors; for starters, the device is activated only with a preliminary cascade (cascades), and only after that connect the final one. In situations where it is technically impossible to turn on without an output transistor, the resistors should be replaced with ones having a nominal value of 5-10 ohms. This will eliminate the possibility of a transistor burnout.
2. Before each restart of the amplifier, it will be necessary to discharge the electrolytic capacitors of the VLF power supply.
3. Check the data on the quiescent current in conditions of low and high temperature of the radiator. The difference in the ratio should be no more than two times. Otherwise, you will have to deal with the ULF thermal stabilizer.

16922

Double-sided JLH2005 amplifier PCB for vintage output transistors in metal cases

Radiators of the driver and current source transistor are tightened with JLH2003 studs for reliability

Installing 2sc5200 output transistors in a JLH 2003 amplifier in plastic cases

The output transistors KT-819 GM, three per shoulder, proved to be no worse than imported ones

Two output transistors and an electronic filter transistor are placed on twisted wires outside the printed circuit board

Budget version of the JLH1969 amplifier on germanium transistors gt404a and mp42b
Selection of output transistors in the amplifier JLH1969 tested kt803

Pre-amplifiers on microcircuits are installed on the JLH2003 termination boards

Circuit boards and case of this JLH2003 amplifier from Chinese online store

The output transistors in the JLH2003 amplifier are soldered directly to the boards.

The class A amplifier of the JLH ideology is assembled according to the scheme - double mono, a flat toroidal transformer is located on the screen

Selection of transistors in the amplifier JLH

Output transistors

In the JLH amplifier, the main attention should be paid to the selection of output transistors in pairs and according to the maximum value of Kus. If you have a very good and easy-to-mount MJL21194, whose Kus is not very high (maximum 50-80), then you need to put a medium power transistor with a beta of at least 150-200 in the driver, for MJ15003 transistors this not so relevant. they have specimens with Kus = 90-120. MJ15003 are more preferable for the output stage due to the parameters, but it is more difficult with them in terms of design. they need to be isolated from radiators.

The input transistor with either those or those transistors must have Kus not less than 250-300. It is not necessary to select transistors for current sources in the 2003 version of the amplifier, although it is also possible to calm the soul. My output transistors are selected with an accuracy of 3-4% and at the same time I didn’t have to pervert especially. I bought obviously original devices, though decently overpaying for them. Of the purchased 16 MJ15003 transistors, their gain spread did not exceed 10-15% at a collector current of 2.5 Amperes. If four (eight) output transistors cannot be selected with an accuracy of 3-5%, then I advise you to put transistors with a large Kus in the lower arm of each amplifier channel (according to the 1969 scheme, this is Tr1). I repeat that the original transistors from the same batch and with the same release date have a beta spread of no more than 15% (IMHO).

Measuring Kus output transistors

Using a multimeter to select powerful transistors by gain is a common mistake. The current at which Kus is measured by industrial multimeters and testers is tens of milliamps, and we need a current approximately equal to the quiescent current in operating mode, i.e. 1.5 - 3 A. The best selection method is immediately after installing the amplifier in the breadboard according to the voltage drop across the resistors included in the emitters of powerful transistors. In addition, in the amplifier layout, the output transistors will warm up to operating temperature, plus full operating current will flow through them. You can select transistors outside the amplifier circuit. To do this, you need to connect the collector of the transistor to the plus of the power supply, and the emitter through a resistor of 0.1-0.3 ohm to the minus. The base of the transistor must be connected through a resistor with a nominal value of 1-2 kOhm to the plus, you can make a circuit from a constant resistor of 0.5 kOhm and a trimming resistor of 1-5 kOhm, then you can change the collector current and calculate the transistor Kus at different values. The transistor must be screwed to the radiators or lowered into a jar of distilled water (we need normal cooling so that the transistor does not warm up above 50-60 degrees). After assembling the circuit, we apply voltage, set the current through the transistor with a trimming resistor in the region of 1.5-2.5 A (the current is controlled by the voltage drop across the resistor 0.1-0.3 Ohm) and let the transistor warm up for about 10-15 minutes. We carry out the same procedure for the rest of the transistors, then we make pairs and quadruples of devices with the closest values ​​​​of the voltage drop across the emitter resistor of 0.1-0.3 Ohm. Such a selection of transistors for JLH will be quite enough.

It is better to measure the base current at fixed values, and select pairs that have close base current at all three measurement points. I used a thick duralumin plate to cool the transistors. I fastened several transistors to it at once and heated the first one before the start of the measurement cycle with a current of 3 A until the temperature of the radiator was fixed at 60 degrees. The remaining transistors took the same temperature and the measurement mode turned out to be close to the actual operating conditions in the final stage.

Collected today one channel of the amplifier. At the input, I put a germanium MP20A with Kus about 70. I soldered a GT404G with Kus 89 into the driver cascade, put KT908A at the output without beta selection. KT908A put on a common radiator with an area of ​​900 sq.cm. through mica pads and paste. After half an hour of warming up, the radiator could be touched, the temperature felt to be about 60 degrees. I really liked the sound to the ear. I don’t know what it is connected with, with 908 at the output or with two germanium ones at the input and driver, but when I assembled the same thing with all silicon transistors, the sound did not convince me at all. Then I tried to replace 908 transistors with KT808, I liked the sound with them less and they warmed up almost instantly. I didn’t have an oscilloscope, so I didn’t understand the reason for the quick warm-up and whether it was excited with the 808s. I tried to change 808 to KT803 and KT-819, and they both work worse than 908, that's for sure. At least for myself, I left them in priority.

Transistors of the USSR = Ostapenko Igor

Good day! As a result of the experiments, I settled on this option: The first transistor is AC125 with Kus 460 (the voice of the entire amplifier depends on this transistor to the maximum). Before AC125, I tried to install the Soviet MP10, 2N3906, BC327 ... these were clearly worse. I tried the Soviet KT801 and KT630d in the driver cascade. With KT630, the amp was excited without a signal, but it sounded better than with imported BD139. KT801 did not like the sound. As a result, I left BD139 with Kus 160 in the driver, and I will still experiment with KT630 ​​and try to remove the excitement. At the exit, I had 100% original TIP3055 and Soviet KT819GM ​​and KT903A with a beta of about 60-80. Imported transistors turned out to be the same in sound as KT903, and KT-819GM ​​remained outsiders. Total: left KT903 for which I had ready-made holes in the radiators. If KT819GM ​​or TIP3055 played better, the radiators would have to be cut.

Now about measurements and sound: I tried to measure the amp through RMAA. It really didn’t work out because my Beringer USB card had higher distortion and its own noise than the amplifier. From which I determined that the noise of the amplifier itself is no more than 90 dB, and the distortion is 0.07% or so. The spectrum is enriched with a dense forest coming from the sound card (. With an amplitude of 22 V at the output, the sinusoid is clean in the range of 20 Hz - 20000 kHz. It turned out about 8 watts at an 8 ohm load. I turned on the amplifier in the broken-down S-90. To be honest, it was I'm surprised... The sound is powerful and thick, so "festive" or something... I haven't heard for a long time that at eight watts in the S-90 woofers were spit out.

Hybrid of JLH1969 and JLH2005 = and4841

I have a device with a unipolar power supply, there is a current source in the driver stage, and the voltage amplifier is powered through a stabilizer on the LM chip. In the output stage, two pairs of 2N3055 matched by Kus (80-90) work. I tried to put 2SC-5200 into the output stage, I didn’t like the sound ... I want to say about the power characteristics. initially did not expect to get a lot of power from JLH without the risk of burning rare imports. The maximum amplitude of each half-wave is almost 16 volts before the top cut. At 4 ohms with a quiescent current of 3 A, the output power reaches 64 watts. This is the peak value, and at this current, the transistors heat up mercilessly, although they are installed on a radiator of about 8000 sq.cm. Now the quiescent current is reduced to 2.1 A, and with it the peak power is about 45 watts, but the transistors work more or less in normal mode. The radiator, for all its monstrosity, cannot cope with heat removal, and four low-speed 120 mm coolers are attached to help it. In each channel there are two CCI transformers with a power of 90 watts each. In total, my amplifier consumes and accordingly dissipates 360 watts in continuous mode. After the transformers, there are two 40 amp diode bridges and filters with a capacity of 3 x 10,000 microfarads per channel. The ground bus is separated by a star from the negative terminals of the filter capacitors. The transistors on the radiators are without gaskets, and the radiators themselves are isolated from the case. To eliminate cotton in the columns, there is a delay circuit.

About transistors in brief:

• Tosiba 1943 and 5200 work well in JLH-59, and for some reason it seemed to me that with direct conduction transistors at the output, the sound is better. When using an "inverted" circuit, there is one plus and one minus in terms of selecting transistors: plus - "good" input n-p-n is a much larger choice (starting from BC239, BC339, 2N2222, 2N3904, 2SC2240 ...); minus - the choice of pre-output p-n-p is much smaller (in principle, only BD140, 2SA1815, 2SB647, 2SB667).
• It is better to assemble a low-power version of the JLH1969 amplifier on import in the 2N3906 driver or the Soviet KT602BM and KT908A outputs at a quiescent current of 1.5 A and a voltage of 12-14 V; and a more powerful one on 2SD667 - 2SD669 or MJE3055T and output MJ15003 with 2.5 A quiescent current and 18-20 V supply. - 1 A.
• Amplifier circuit with bipolar power supply and modern details: Output stage on 2sc5200, pre-output stage - BD137 Philips and BD139 Fairchild, 2SC3421 (2SC5171 pleased with detail), input low-noise - 2SA970 (BL) and BC560 (C), current source transistors - MPSA56 / 92 ... sounds very interesting, the harmonics are limited to the 3rd and there is very little of it. Measured at 30 kHz.
• In both versions of the amplifier there is no RF correction, therefore, when using RF transistors, self-excitation is possible and many advise using low-frequency transistors. But low-frequency transistors fill up the meander front, everything is much better with high-frequency transistors, you need to apply a correction with them, and the frequency of the first pole should be more than 25 kHz, because at the pole below 20-25 kHz, the blockage is clearly audible at the top.
• On the sound, there is a strong difference between the inverting and non-inverting versions of the amplifier (those parallel and serial OOS). The difference between the schemes of 1969 and 2005 is not so great, although, as for me, the 1969 is more pleasant. For the 1969 circuit, with 2sc5200 transistors at the output, in parallel with the OOS resistor going from the output to the emitter of the first transistor, you need to put a capacitor with a capacity of 33-68 pF (when this resistor is halved to 1.2 kOhm, the capacitance of this capacitor must be increased to 47-100 pF). The second correction element is the capacitance between the collector and the base of the pre-output transistor, set 6-15 pF, and if you reduce the value of the resistor in the collector of the first stage to 4 kOhm, then 10-27 pF. This capacitance should be chosen as minimum in the absence of excitation. The only problem with the inverting circuit is that its input resistance is constant and equal to the value of the input resistor (1 kΩ in the circuit), which means that a non-standard low-resistance volume control with a value of less than 1 kΩ is needed here. Plus, the inverting circuit imposes a hard limit on the output impedance of the signal source, which should not exceed hundreds of ohms. In the inverse connection, the sound is much better and the input transistor works with OB (less distortion). By far the best sound I've heard = FEDGEN
• Of the transistors for use in the output stage, I have not seen a better MJ15024 / MJ15025, it’s generally a disaster with pre-output ones. You can try Tosiba 2SA1302\2SC3281, 2SA1987\2SC5359, they are more stable and complimentary = Vlad Bo.
• Problems in modern transistors - what not to do with them in the HF region, there is squeakiness, especially SANKENs, and in LAPTs (multi-emitter). I love Motorola MJ15025, on Japanese amplifiers that I came across, I replaced all Japanese ones with Motorola. Transistors MJ15025 ideal for sound in terms of frequency properties are not yet the best. Yes, and by ear Motorola MJE15003, MJE15004 sound better than Toshiba - 2sc5200, 2sc1943.

P. S. Whoever assembled this device is praised. Especially using old Motorolas or our old germanium. If we implement the scheme