Lanzar amplifier on domestic parts. Setting up a Lanzar power amplifier - a circuit diagram of a power amplifier, a description of the circuit diagram, recommendations for assembly and adjustment. Other useful information and possible troubleshooting

WE COLLECT LANZAR

The repetition of the same questions on every page of this amp's discussion prompted me to write this little sketch. Everything written below is my idea of ​​what you need to know. beginner to a radio amateur who decided to make this amplifier, and does not claim to be an absolute truth.

Let's say you are looking for a good transistor amplifier circuit. Such schemes, such as "UM Zueva", "VP", "Natalie", and others seem complicated to you, or there is little experience to assemble them, but good sound I want to. Then you have found what you were looking for! Lanzar is an amplifier built according to the classical symmetrical scheme, with an output stage operating in class AB, and has a pretty good sound, in the absence of complex settings and scarce components.

Amplifier circuit:

I found it necessary to make some minor changes to the original circuit: the gain is slightly increased - up to 28 times (changed R14), the values ​​\u200b\u200bare changed input filter R1, R2 and also on advice MayBe I'm a Leo resistor values ​​of the base divider of the thermal stabilization transistor (R15, R15’) for more smooth tuning rest current. The changes are not critical. The numbering of the elements has been preserved.

Amplifier Power

Amplifier power supply- the most expensive link in it, so you should start with it. Below are a few words about IP.

Based on the load resistance and the desired output power, the desired supply voltage is selected (Table 1). This table taken from the source site (interlavka.narod.ru), but, I personally strongly I would not recommend operating this amplifier at powers of more than 200-220 watts.

REMEMBER! This is not a computer, no super-cooling is needed, the design should not work at the limit of its capabilities, then you will get a reliable amplifier that will work for many years and delight you with sound. After all, we decided to make a high-quality device, and not a bouquet of New Year's fireworks, so let all sorts of "squeezers" go through the forest.

At supply voltages below ±45 V / 8 ohms and ± 35 V / 4 ohms, the second pair of output transistors (VT12, VT13) can be omitted! With such supply voltages, we get an output power of about 100 W, which is more than enough for a house. I note that if at such voltages you still install 2 pairs, then the output power will increase by a completely insignificant amount of about 3-5 watts. But if "the toad does not choke", then in order to increase reliability, you can put 2 pairs.

Transformer power can be calculated using the program powerup. Calculation based on the fact that the approximate efficiency of the amplifier is 50-55%, which means that the power of the transformer is: Ptrans \u003d (Pout * N channels * 100%) / efficiency applicable only if you want to listen to a sinusoid for a long time. A real musical signal, unlike a sine wave, has a much smaller peak-to-average ratio, so it makes no sense to spend money on extra transformer power that will never be used anyway.

In the calculation, I recommend choosing the heaviest peak factor (8 dB) so that your PSU does not bend if you suddenly decide to listen to music with such a pf. By the way, I also recommend calculating the output power and supply voltage using this program. For Lanzar dU, you can choose about 4-7 V.

More about the program powerup and the calculation method is written on site author (AudioKiller).

All this is especially true if you decide to buy a new transformer. If you already have it in the bins, and suddenly it turned out to be more powerful than the calculated one, then you can safely use it, the stock is a good thing, but fanaticism is not needed. If you decide to make a transformer yourself, then on this page of Sergey Komarov there is a normal calculation method .

The schema itself the simplest bipolar power supply looks like that:

The scheme itself and the details for its construction are well described by Mikhail (D-Evil) in FAKe according to TDA7294.

I will not repeat myself, I will only note the amendment about the power of the transformer, described above, and about diode bridge: since Lanzar's supply voltage can be higher than TDA729x, the bridge must "hold" a correspondingly larger reverse voltage, not less than:

Urev_min = 1.2*(1.4*2*Uhalf-winding_transformer) ,

where 1.2 is the safety factor (20%)

And at high transformer powers and capacities in the filter, in order to protect the transformer and the bridge from colossal inrush currents, the so-called. "soft start" or "soft start" scheme.

Amplifier details

A list of parts for one channel is attached in the archive in

Some denominations require special explanation:

C1- isolation capacitor, must be of good quality. There are different opinions on the types of capacitors used as separators, so the sophisticated will be able to choose the best option for themselves. For the rest, I recommend using polypropylene film capacitors of well-known brands such as Reef PHE426, etc., but in the absence of such, widely available lavsan K73-17 are quite suitable.

The lower cutoff frequency, which will be amplified, also depends on the capacitance of this capacitor.

In the printed circuit board from interlavka.narod.ru, as C1, a seat is provided for a non-polar capacitor, composed of two electrolytes, included by "minuses" to each other and "pluses" in the circuit and shunted by a 1 μF film capacitor:

Personally, I would throw out the electrolytes and leave one film capacitor above the indicated types, with a capacity of 1.5-3.3 microfarads - this capacity is enough for the amplifier to operate on a "wide band". In the case of working on a subwoofer, more capacity is required. Here it would be possible to add electrolytes with capacities of 22-50 microfarads x 25 V. However, the printed circuit board imposes its limitations, and a 2.2-3.3 microfarad film capacitor is unlikely to fit there. Therefore, we set 2x22 microfarads 25 V + 1 microfarads.

R3, R6- ballast. Although initially these resistors were selected as 2.7 kOhm, I would recalculate them for the desired amplifier supply voltage using the formula:

R \u003d (Ushoulder - 15V) / Ist (kOhm) ,

where Ist - stabilization current, mA (about 8-10 mA)

L1- 10 turns of 0.8 mm wire on a 12 mm mandrel, everything is smeared with superglue, and after drying, a resistor is inserted inside R31.

Electrolytic Capacitors C8, C11, C16, C17 must be designed for a voltage not lower than the supply voltage with a margin of 15-20%, for example, at ± 35 V, 50 V capacitors are suitable, and at ± 50 V, you already need to choose 63 Volts. The voltages of other electrolytic capacitors are indicated in the diagram.

Film capacitors (non-polarized) are usually not made rated for less than 63 V, so there shouldn't be a problem.

Trimmer resistor R15– multi-turn, type 3296.

Under emitter resistorsR26, R27, R29 and R30– the board provides seats for wire ceramic SQP resistors with a power of 5 watts. The range of acceptable ratings is 0.22-0.33 ohms. Although SQP is far from the best option, it is affordable.

Domestic resistors C5-16 can also be used. I haven't tried it, but they might even be better than SQP.

Other resistors- C1-4 (carbon) or C2-23 (MLT) (metal film). All, except those indicated separately - at 0.25 watts.

Some possible substitutions:

1. Paired transistors are changed to other pairs. Making a pair of transistors of two different pairs is unacceptable.
2. VT5/VT6 can be replaced by 2SB649/2SD669. It should be noted that the pinout of these transistors is mirrored relative to 2SA1837 / 2SC4793, and when used, they must be rotated 180 degrees relative to those drawn on the board.
3. VT8/VT9– on 2SC5171/2SA1930
4. VT7– on BD135, BD137
5. Diffcascade transistors ( VT1 andVT3), (VT2 andVT4) it is desirable to select in pairs with the smallest spread of beta (hFE) using a tester. An accuracy of 10-15% is enough. With a strong spread, a slightly increased level of constant voltage at the output is possible. The process is described by Mikhail (D-Evil) in the FAQ on the VP amplifier .

Another illustration of the beta measurement process:

The 2SC5200/2SA1943 transistors are the most expensive components in this circuit and are often counterfeited. Similar to the real 2SC5200 / 2SA1943 from Toshiba, they have two break marks on top and look like this:

It is advisable to take the same output transistors from the same batch (in Figure 512 - the batch number, that is, let's say both 2SC5200 with the number 512), then the quiescent current when installing two pairs will be more evenly distributed to each pair.

Printed circuit board

The printed circuit board was taken from interlavka.narod.ru. The corrections on my part were mostly cosmetic in nature, some errors in the signed denominations were also corrected, such as mixed up resistors for the thermal stabilization transistor and other little things. The board is drawn from the side of the details. You don't need to mirror to make a LUT!

1. IMPORTANT! Before by soldering each the part must be checked for serviceability, the resistance of the resistors is measured to avoid an error in the nominal value, the transistors are checked by a continuity tester, and so on. Searching for such errors later on the assembled board is much more difficult, so it's better not to rush and check everything. save a bunch time and nerves.
2. IMPORTANT! Before soldering the trimmer resistor R15, it must be “unscrewed” so that its impedance is soldered into the gap of the track, i.e., if you look at the picture above, between the right and middle output should be. all the resistance of the trimmer.
3. Jumpers to prevent accidental short circuit it is better to do it with insulated wires.
4. transistors VT7-VT13 are installed on a common radiator through insulating gaskets - mica with thermal paste (for example, KPT-8) or Nomacon. Mica is more preferable. Indicated on the diagram VT8, VT9 in an insulated case, so their flanges can simply be lubricated with thermal grease. After installation on the radiator, the tester checks the transistor collectors (middle legs) for the absence of a short circuit. with radiator.
5. transistors VT5, VT6 also need to be installed on small radiators - for example, 2 flat plates about 7x3 cm in size, in general, put what you can find in the bins, don’t forget to just grease with thermal paste.
6. For better thermal contact, differential stage transistors ( VT1 and VT3), (VT2 and VT4) can also be lubricated with thermal paste and pressed against each other with heat shrink.

First run and setup

Once again, we carefully check everything, if everything looks fine, there are no errors anywhere, “snot”, short circuits to the radiator, etc., then you can proceed to the first launch.

IMPORTANT! The first start-up and configuration of any amplifier must be carried out with input shorted to ground, with power supply current limiting and no load . Then the chance to burn something is greatly reduced. The simplest solution I use is incandescent lamp 60-150 W, connected in series with the primary winding of the transformer:

We start the amplifier through the lamp, measure the constant voltage at the output: normal values ​​\u200b\u200bare no more than ± (50-70) mV. "Walking" constant within ± 10 mV is considered normal. We control the presence of voltages of 15 V on both zener diodes. If everything is normal, nothing exploded, did not burn out, then we proceed to the setup.

When starting a working amplifier with a quiescent current = 0, the lamp should flash for a short time (due to the current when the capacitances are charged in the PSU), and then go out. If the lamp burns brightly, then something is faulty, turn it off and look for an error.

As already mentioned, the amplifier is easy to set up: you only need set quiescent current (TC) output transistors.

It should be displayed on "warmed up" amplifier, i.e. before installation, let it play for a while, 15-20 minutes. During the installation of the transformer substation, the input must be shorted to ground, and the output should hang in the air.

The quiescent current can be found by measuring the voltage drop across a pair of emitter resistors, such as R26 and R27(set the multimeter to the limit of 200 mV, the probes to the emitters VT10 and VT11):

Accordingly, Ipok \u003d Uv / (R26 + R26) .

Further SMOOTHLY, without jerks, turn the trimmer and look at the readings of the multimeter. Required to install 70-100 mA. For the resistor values ​​shown in the figure, this is equivalent to a multimeter reading of (30-44) mV.

The light bulb may start to glow a little. We check once again the level of constant voltage at the output, if everything is normal, you can connect acoustics and listen.

Photo of the assembled amplifier

Other useful information and possible troubleshooting options

Amplifier self-excitation: Indirectly determined by the heating of the resistor in the Zobel circuit - R28. Reliably determined using an oscilloscope. To eliminate, try to increase the values ​​​​of corrective capacitances C9 and C10.

Large level of the DC component at the output: pick up differential cascade transistors ( VT1 and VT3), (VT2 and VT4) according to "Betta". If it doesn’t help, or it’s not possible to choose more precisely, then you can try changing the value of one of the resistors R4 and R5. But such a solution is not the best, it is still better to choose transistors.

Option for a slight increase in sensitivity: You can increase the sensitivity of the amplifier (gain) by increasing the value of the resistor R14. Coef. gain can be calculated by the formula:

Ku \u003d 1 + R14 / R11, (once)

But do not get too carried away, as with an increase R14, the depth of the FOS decreases and the unevenness of the frequency response and SOI increases. It is better to measure the output voltage level of the source at full volume (amplitude) and calculate what Ku is needed for the amplifier to operate at full output voltage swing, taking it with a margin of 3 dB (before clipping).

For specifics, let the maximum to which it is tolerable to raise Ku - 40-50. If you need more, then make a preamplifier.

If you have any questions, write in the appropriate topic to the forum . Happy assembly!

ULF Lanzar (Lanzar) is an amplifier built according to the classical symmetrical circuit, operating in class AB. A lot of car amplifiers are assembled in a similar way. simple circuit, the "fluffiness" of the assembly and tuning of this amplifier on numerous forums is a guarantee of success for novice amp builders. It is enough that the hands grow from the right places, all that remains is to solder everything correctly and set the quiescent current, that's the whole setting. Therefore, after assembling amplifiers on microcircuits (TDA7294), Lanzar may well serve as the next step. The sound is quite decent, unpretentious and hardy, can be used to work with subwoofers. As output transistors, you can use bipolar and FETs.

Scheme ULF Lanzar

Even from Interlavka, it was customary to make Lanzars according to such a layout. Uh, in light of the latest trends in PCB wiring, it's just awful ...

The contours of the power and ground buses are very long, and the power conductors are thin, it is necessary to breed exactly the opposite. Although once upon a time my first collected and earned ULF was Lanzar with all these shortcomings). And then I had some progress in mastering PCB layout in P-CAD, taking into account the recommendations on the forums. It turned out like this Lanzar on the fields, the PP is double-sided, the top layer is mostly green in the form of a continuous polygon. It turned out compact and according to Feng Shui)

Board layout on bipolars with one pair at the output:

First, we check the correctness of the wiring with the LUT, otherwise you will miss the jamb and it will multiply when ordering PP in production ... This is how ULF Lanzar looks assembled on one pair of bipolars. The PP is double-sided, I had to wrinkle it with an iron, aligning the printouts with pins at the control points. In general, it turned out fine and the channels launched immediately.

Since there were no errors in the wiring, you can also order PP at the production site, because. the series has not been planned yet, then to save money without a mask and markings:

The question is regularly asked: “How to wind output coil". Simple: we take a drill (mandrel) with a diameter of 5.7-5.8 mm, an enamel wire 1-1.1 mm, wind 8 turns back and forth and 7 back. We clean, mold after landing, everything is ready.

Lanzar also spread two pairs of bipolars, soldered and launched from a half turn:

The photo was preserved only without the endings, because. did not have time to solder, the amplifier "found" a new owner)

It is wound on a 10mm drill and consists of 10 turns of 0.8mm wire; for rigid fixation of the turns, you can spread superglue on the finished coil.

The emitter resistors of the output transistors are selected with a power of 5 watts, during operation they overheat. The value of these resistors is not critical and can be from 0.22 to 0.39 ohms.

After the assembly of the amplifier is completed, we proceed to the verification stage. We carefully call the conclusions of the transistors and check for short circuits 0 they should not be. Then we look at the installation again, check the board by eye - we pay special attention to the correct connection of transistors and zener diodes, if some transistors were replaced with similar ones, then see the manuals, since the conclusions of the transistors and analogues used in the circuit may differ.

The zener diodes themselves, if connected incorrectly, work like a diode and there is a chance of ruining the entire circuit due to an incorrectly connected zener diode.

Variable resistor to adjust the quiescent current of the output stages - it is advisable to use (very even desirable) multi-turn resistors with a resistance of 1 kOhm, while the resistance during installation should be maximum - 1 kOhm. A multi-turn resistor will allow you to adjust the quiescent current of the output stage with very high accuracy.

It is advisable to take all electrolytic capacitors with an operating voltage of 63, and even better, 100 volts.

Before assembling the amplifier, we carefully check all components for serviceability, regardless of whether they are new or used.

Good evening, gentlemen radio amateurs! It all started with the fact that in home UMZCH I have long wanted to abandon cheap TDA-niks and switch to more high level- a decent transistor audio frequency amplifier. I read many pages of a wide variety of forums, looked through various photo galleries, reviewed reviews ... and decided to try to assemble a novelty for myself, the choice fell on a very well-known and good Lanzar amplifier in terms of characteristics. The next month was spent studying all possible varieties of circuits for this amplifier and choosing the optimal and suitable one in terms of characteristics.

Schematic diagram of ULF Lanzar

It seemed to me relatively easy to repeat and customize, although it is she who is given the most attention in all the forums! Well, I went to the radio market, bought parts, at a price it cost me 110 UAH - a lot, as for a student, I’ll tell you, but the end result was worth it, more on that later ... I took up manufacturing printed circuit board, with etching it took an hour and a half. I poisoned with ferric chloride, I have not yet got used to it, since I mainly use copper sulphate. After preparing the board of the future, Lanzara took up soldering, first of all, jumpers were soldered, then resistors, capacitors, transistors ...

Having soldered the board, we proceed to the main thing - setting the idling current of the UMZCH. Here everything was simple for me - I set the trimmer to the average value, soldered it, checked the board for snot and turned it on. Even without fuses (not like light bulbs). Lanzar started up immediately, drove it for 15 minutes until the VC heated up, but the trimmer did not pull, measured the voltage drop across five-watt resistors - it did not change, no noise, and no other noticeable distortions were found with an oscilloscope, which showed a high repeatability of this circuit!

Now about the sound impressions: earlier when listening to tda7294 for at least an hour and the subsequent exception, there was a feeling as if a tightly pulled helmet was removed from the head, then I realized that this was due to the lack of mid frequencies in tda7294 .

Now it's the turn to load the lanzar with a pair of low-power speakers, since I have a test power of + -22V, then small 25-watt speakers were just right for it.

Photo of the finished UMZCH

As you can see from the pictures, the power capacitors are not very fat in terms of power supply, only 470 microfarads, but in terms of voltage they are with a large margin, since it is planned to power Lanzar from + - 65V in the future! Such speakers were connected to the amplifier during the tuning process.

Having a powerful, high-quality subwoofer is the desire of every car enthusiast who appreciates high-quality and loud sound and deep low frequencies(bass). The project was implemented in the summer of 2012 and took as much as 3 months, such a delay due to the lack of many components that were used in the project. The device itself is a complex of amplifiers with a total power of about 750-800 watts. In several articles, I will try to explain in detail the design of the Lanzar subwoofer amplifier.

A voltage converter, a filter adder, a stabilizer block and a dynamic head protection are the components for the operation of such an amplifier. The voltage converter develops a power of 500 watts, and all these 500 watts are directed to power the main amplifier. Lanzar power can reach up to 360-390 watts, although the maximum power is obtained with increased nutrition and dangerous enough for individual parts of the amplifier.

Such an amplifier feeds a powerful self-made subwoofer based on a SONY XPLOD dynamic head with a rated power of 300-350 watts, maximum (short-term power) up to 1000 watts. In a separate article, we will consider the process of manufacturing a box for a subwoofer and all the subtleties associated with it. The case was used from a DVD player, it fit perfectly in size. To cool the main amplifier, a huge heat sink from a Soviet radio engineering amplifier was used. There is also a high-speed laptop cooler to get warm air out of the case.

Let's start considering the design with a voltage converter, since it will be necessary to do it in the first place. The entire operation of the structure depends on the precise operation of the converter. It provides a bipolar voltage of 60 volts per arm at the output, which is exactly what is needed to provide the specified output power of the amplifier.

Voltage converter, despite simple design develops a power of 500 watts, in force majeure situations up to 650 watts. TL494 - two-channel PWM controller, generator rectangular pulses tuned to a frequency of 45-50 kHz is the engine of this converter, it all starts with it.

To amplify the output signal, a driver is assembled on low-power bipolar transistors BC556(557) series.

Pre amplified signal through limiting resistors is fed to the gates of powerful power switches. This circuit uses powerful N-channel field-effect transistors of the IRF3205 series, there are 4 of them in the circuit.

The converter transformer was initially wound on two cores (W-shaped) from the block ATX power supply, but then the design changed, and a new transformer was wound. Ring from an electronic transformer for powering halogen lamps (power 150-230 watts). The transformer contains two windings. The primary winding is wound immediately with 10 strands of wire 0.5-0.7 mm and contains 2x5 turns. Winding is done like this. To begin with, we take a test wire and wind 5 turns, stretch the turns around the entire ring. We unwind the wire and measure its length. We make measurements with a margin of 5 cm. Then we take 10 cores of the same wire - we twist the ends of the wires. We make two such blanks - 2 tires of 10 cores. Then we try to wind as evenly as possible around the entire ring, we get 5 turns. Then you need to separate the tires, as a result we get two equivalent halves of the winding.

We connect the beginning of one winding with the end of the second winding, or vice versa - the end of the first with the beginning of the second. Thus, we have phased the windings and the circuit can be checked. To do this, we connect the transformer to the circuit, and wind the test winding (secondary) on the ring. The winding can contain any number of turns, it is better to wind 2-6 turns of wire 0.5-1mm.
The first start of the converter is best done through a lamp (halogen) at 20-60 watts.

After winding the trial secondary winding, we start the converter. We connect an incandescent lamp with a power of a couple of watts to the test winding. The lamp should glow, while the transistors (if so far without heat sinks) should heat up slightly during operation.
If everything is fine, then you can wind a real winding, if the circuit does not work properly or does not work at all, then you need to turn off the gates of the transistors and use an oscilloscope to check for rectangular pulses at pins 9 and 10. If there is generation, then the problem is most likely in the transistors, if they are also normal, then the transformer is incorrectly phased, you need to change the beginning and end of the windings (phasing was discussed in the 2nd part).

The secondary winding is wound according to the same principle as the primary, it is also phased in the same way. The winding contains 2X18 turns and is wound immediately with 8 wires of 0.5 mm wire. The winding must be stretched over the entire ring. The center tap will be the case, since we need to get a bipolar voltage. The output voltage is obtained with an increased frequency, so the multimeter is not able to measure it.
The diode rectifier in my case was assembled from powerful domestic diodes of the KD213A series. The reverse voltage of the diode is 200V, at a current of up to 10A, These diodes can operate at frequencies up to 100kHz - a great option for our case. You can also use other powerful pulse diodes with a reverse voltage of at least 180 volts.