Switching power supply for low-frequency power amplifier. Switching power supply for UCH Switching power supply for UCH
After successful, we move on to the most interesting part of the design - the block of audio power amplifiers. Including a low-pass filter for the subwoofer and a stabilization module. We remind you that all circuit diagrams and drawings of boards are .Well, what can we say about one of the most repeated power amplifier circuits - the Lanzar circuit was developed back in the 70s of the last century. On a modern high-precision elementary base, Lanzar began to sound even better. In theory, the circuit is excellent for wideband acoustics, distortion at half the volume only 0.04%- full-fledged Hi-Fi.
The output stage of the amplifier is built on a pair 2SA1943 And 2SC5200, all stages are assembled on complementary pairs that are as close as possible in terms of parameters, the amplifier is built entirely on a symmetrical basis. The rated output power of the amplifier is 230-280 watts, but much more can be removed by increasing the input supply voltage.
The values of the limiting resistors of the differential stages are selected based on the input voltage. Below is the table.
Power supply ±70 V - 3.3 kOhm...3.9 kOhm
Power supply ±60 V - 2.7 kOhm...3.3 kOhm
Power supply ±50 V - 2.2 kOhm...2.7 kOhm
Power supply ±40 V - 1.5 kOhm...2.2 kOhm
Power supply ±30 V - 1.0 kOhm...1.5 kOhm
These resistors are selected with a power of 1-2 watts; during operation, heat generation may be observed on them.
The regulating transistor was replaced with a domestic one KT815, at that time there was no other one at hand. It is designed to regulate the quiescent current of the output stages; it does not overheat during operation, but is mounted on a common heat sink with the transistors of the output stage.
It is advisable to do the first start of the circuit from the mains power supply; connect a 100-150 watt incandescent lamp in series with the mains winding of the transformer; if there are problems, then burn a minimum of parts. In general, Lanzar’s circuit is not critical to the installation and components; I tried it even with a wide range of components used, using domestic radio components - the circuit shows high parameters even in this case. Lanzar's circuit diagram has two main versions - on bipolar transistors and using field switches in the penultimate stage, in my case first version.
The second pre-output stage operates in a pure class " A", therefore, during operation, the transistors overheat. Transistors of this cascade must be installed on a heat sink, preferably a common one, do not forget about insulation - mica plates and insulating washers for screws.
A correctly assembled circuit starts up without any problems. We do the first launch with INPUT SHORT TO GROUND
, i.e. The amplifier input is connected to the middle point from the power supply. If nothing explodes after launch, then you can disconnect the input from the ground. Next we connect the load - the speaker and turn on the amplifier. To make sure the amplifier is working, just touch the bare input wire. If a peculiar roar appears in the head, then the amplifier is working! Next, you can strengthen all the power parts with heat sinks and send an audio signal to the amplifier input. After 15-20 minutes of operation at 30-50% of the maximum volume, you need to adjust the quiescent current. The photo shows everything in detail; it is advisable to use a digital multimeter as a voltage indicator.
Amplifier output power measurement
How to set the quiescent current
The low-pass filter and adder are built on two microcircuits. It is designed for smooth adjustment of phase, volume and frequency. The adder is designed to sum the signals of both channels to obtain a more powerful signal. Industrial high-power auto amplifiers use exactly this principle of filtering and summing the signal, but the adder can, if desired, be excluded from the circuit and use only a low-pass filter. The filter cuts off all frequencies, leaving only a limit between 35-150 Hz.
Phase adjustment allows you to match the subwoofer with speaker systems; in some cases it is also excluded.
This unit is powered by a stabilized bipolar voltage source +/-15 Volts. Power can be provided using an additional secondary winding, or you can use a bipolar voltage stabilizer to reduce the voltage from the main winding.
For this purpose, a bipolar stabilizer has been assembled. Initially, the voltage is reduced by zener diodes, then amplified by bipolar transistors and supplied to linear voltage regulators such as 7815
And 7915
. At the output of the stabilizer, a stable bipolar power supply is formed, which powers the adder and low-pass filter unit.
Stabilizers and transistors can get hot, but this is quite normal; if desired, they can be mounted on heat sinks, but in my case there is active cooling by a cooler, so heat sinks were not useful, and besides, the heat dissipation is within normal limits, since the low-pass filter unit itself consumes very little.
SLAPPING TO ICROCHIRECTS
Slap in the face mikruham is not the simplest, but high-quality low-frequency power amplifier. The amplifier is capable of developing a maximum output power of 130 watts and operates over a fairly wide input voltage range. The output stage of the amplifier is built on a pair 2sa1943 2sc5200 and works in mode AB. This version was developed by the author this year, below are its main parameters. Supply voltage range = +/- 20V... +/- 60V
Nominal supply voltage (100W, 4 Ohm) = +/- 36V
Nominal supply voltage (100W, 8 Ohm) = +/- 48V
Everything is clear with power, but what about distortion?
THD+N (at Pout<=60Вт, 20кГц) <= 0,0009%
THD+N (at maximum output power, 20kHz) = 0.008%
The parts used in this module are trimming resistors, low- and medium-power transistors:
HERE VIDEO
Not bad at all, almost high-end! In fact, if you focus only on SOI, then this amplifier is full-fledged HI-END, but this is not enough for high-end, so it was classified in the good old category hi-fi.
Although the amplifier develops only 100 watts, it is an order of magnitude more complex than similar circuits, but the assembly itself will not be difficult if all the components are available. I do not recommend rejecting the circuit values - my experience confirms this.
Low-power transistors may overheat during operation, but there is no need to worry - this is their normal operating mode. The output stage, as already said, operates in class AB, therefore, a huge amount of heat will be released that needs to be removed. In my case, they are reinforced with a common heat sink, which is more than enough, but just in case, there is also active cooling.
After assembly, we are waiting for the first launch of the circuit. To do this, I advise you to read again about launching and configuring Lanzar - here everything is done in exactly the same way. We do the first start with the input shorted to ground, if everything is OK, then we open the input and sound a sound signal. By that time, all power components must be strengthened with a heat sink, otherwise, while admiring the music, you may not notice how the output stage switches smoke - each of them is very, very expensive. And you will learn about the protection unit. Sincerely - AKA KASYAN.
Discuss the article HOME AMPLIFIER - UMZCH UNIT
Other articles dedicated to the construction of this ULF.
Schematic diagram of the power supply.
The power supply is assembled according to one of the standard schemes. Bipolar power supply is selected to power the final amplifiers. This allows the use of inexpensive, high-quality integrated amplifiers and eliminates a number of problems associated with supply voltage ripple and turn-on transients. https://site/
The power supply must provide power to three microcircuits and one LED. Two TDA2030 microcircuits are used as final power amplifiers, and one TDA1524A microcircuit is used as a volume control, network base and tone.
Electrical diagram of the power supply.
|
VD3... VD6 – KD226 |
C1 – 680mkFx25V C3... C6 – 1000mkFx25V |
A bipolar, full-wave rectifier with a midpoint is assembled using diodes VD3...VD6. This connection circuit reduces the voltage drop across the rectifier diodes by half compared to a conventional bridge rectifier, since in each half-cycle the current flows through only one diode.
Electrolytic capacitors C3...C6 are used as a rectified voltage filter.
The IC1 chip contains a voltage stabilizer to power the electronic volume, stereo and tone control circuits. The stabilizer is assembled according to a standard design.
The use of the LM317 chip is due only to the fact that it was available. Here you can use any integral stabilizer.
The protective diode VD2, indicated by a dotted line, is not necessary to use when the output voltage on the LM317 chip is below 25 Volts. But, if the input voltage of the microcircuit is 25 Volts or higher, and resistor R3 is a tuning resistor, then it is better to install a diode.
The value of resistor R3 determines the output voltage of the stabilizer. During prototyping, I soldered a trimmer resistor in its place, used it to set the voltage to about 9 Volts at the output of the stabilizer, and then measured the resistance of this trimmer so that I could install a constant resistor instead.
The rectifier feeding the stabilizer is made according to a simplified half-wave circuit, which is dictated by purely economic considerations. Four diodes and one capacitor are more expensive than one diode and one slightly larger capacitor.
The current consumed by the TDA1524A microcircuit is only 35mA, so this circuit is quite justified.
LED HL1 – amplifier power-on indicator. A ballast resistor for this indicator is installed on the power supply board - R1 with a nominal resistance of 500 Ohms. The LED current depends on the resistance of this resistor. I used a green LED rated at 20mA. When using a red LED type AL307 with a current of 5mA, the resistance of the resistor can be increased by 3-4 times.
Printed circuit board.
The printed circuit board (PCB) is designed based on the design of a specific amplifier and available electrical components. The board has only one hole for mounting, located in the very center of the PCB, which is due to its unusual design.
To increase the cross-section of copper tracks and save ferric chloride, the areas free from tracks on the PP were filled using the “Polygon” tool.
Increasing the width of the tracks also prevents peeling of the foil from the fiberglass laminate in the event of a violation of the thermal regime or during repeated re-soldering of radio components.
According to the drawing given above, a printed circuit board was made from foil fiberglass with a cross section of 1 mm.
To connect the wires to the printed circuit board, copper pins (soldiers) were riveted into the holes of the board.
This movie requires Flash Player 9 |
||
And this is the already assembled printed circuit board of the power supply.
To see all six views, drag the picture with the cursor or use the arrow buttons located at the bottom of the picture.
The mesh on the PP copper tracks is the result of using this technology. 
When the board is assembled, it is advisable to test it before connecting the final amplifiers and the regulator unit. To test the power supply, you need to connect an equivalent load to its outputs, as in the diagram above.
Resistors of the PEV-10 type at 10-15 Ohms are suitable as a load for the +12.8 and -12.8 Volt rectifiers.
It’s a good idea to look at the voltage at the output of a stabilizer loaded onto a resistor with a resistance of 100-150 Ohms with an oscilloscope to ensure there is no ripple when the alternating input voltage decreases from 14.3 to 10 Volts.
P.S. Refinement of the printed circuit board.
During commissioning, the printed circuit board of the power supply was damaged.
During modification, we had to cut one track, item 1, and add one contact, item 2, to connect the transformer winding that powers the voltage stabilizer.
If you need a power supply for non-standard conditions, you can use a design with a low-frequency transformer. This solution is easy to implement and does not require particularly deep special knowledge, but it also has a number of disadvantages - large dimensions, low efficiency and quality of output voltage stabilization. It is possible to make a switching power supply, but this is a rather complicated procedure with a lot of pitfalls - with the slightest mistake there will be a “pop” and a bunch of unnecessary parts.
Let's try to lower the bar and limit ourselves to upgrading a conventional ATX computer power supply to meet the necessary requirements. Hm, what exactly will be the subject of consideration? In fact, a 300-400 watt power supply can provide quite significant power; its scope of application is large. It is difficult to cover the immensity in one article, so we will limit ourselves to the most common one - a low-frequency amplifier, and we will try to carry out a modification for it.
Formulation of the problem
The power supply is quite powerful, I would like to use it to the maximum. You cannot make a powerful amplifier from 12 volts; a completely different approach is required here - bipolar power supply with an output voltage clearly greater than 12 V. If the power supply will power a homemade amplifier assembled from discrete elements, then its supply voltage can be any (within reasonable limits), and Integrated circuits are quite picky. To be specific, let’s take an amplifier with a supply voltage of up to 100 V (+/-50 V) with an output power of 100 W. The microcircuit provides a dynamic current of up to 10 amperes, which determines the maximum load current of the power supply.
Everything seems clear, all that remains is to clarify the output voltage level. Operation from a power supply of 100 volts (+/-50 V) is acceptable, but attempting to select such an output voltage would be a big mistake. Microcircuits have an extremely negative attitude towards extreme operating modes, especially when several parameters are simultaneously at maximum values - supply voltage and power. Moreover, it hardly makes sense in an ordinary apartment to provide such a high level of power, even for low-frequency speakers with their low efficiency.
It is possible to set the voltage to 90 volts (+/- 45 V), but this would require very precise holding of the output voltage - in multi-channel power supplies it is very difficult to ensure the same voltages at different outputs. Therefore, it is worth lowering the bar a little and setting the nominal voltage for this microcircuit to 80 volts (+/-40 V) - the amplifier power will drop slightly, but the device will work with the proper safety margin, which will ensure sufficient reliability of the device.
In addition, if the speaker will work not only in the low-frequency region, but also contains mid-high-frequency amplifier channels, then it is worth getting another voltage from the power supply, less than “+/-40 V”. The operating efficiency of large-diameter low-frequency speakers is significantly lower than higher-frequency ones, so powering the amplifier of the mid-high-frequency channel from the same “+/-40 V” is quite stupid, the bulk of the energy will go into heat. For the second amplifier it would be good to provide an output of +/-20 volts.
So, the specification of the power supply that you want to get:
- Channel No. 1 (main), voltage: “+/-40 V”.
- Load current from 0.1 A to 10 A.
- Channel No. 2 (additional), voltage: “+/-20 V”.
- Load current from 0 to 5 A.
The characteristics have been determined, all that remains is to choose the appropriate model. There is no desire to use a completely old one, the capacitors have long since dried out, and the circuit solutions of those times do not inspire optimism. It is worth noting that some of the “modern” power supplies also do not shine with quality of work and reliability, but you can fight this - just choose products from well-known companies that you can trust.
In addition to the philosophical understanding of the essence of BP and selection by appearance, there is a completely meaningful criterion - their type. The block can be made using the “push-pull half-bridge” or “single-stroke forward” technology, and contain some kind of PFC (active or passive on the throttle). All these factors influence the quality of work and the level of interference. Moreover, these are not “just words”; when switching from a transformer power supply to a “pulse” one, a deterioration in sound quality is often noticed.
On the one hand, it’s “strange”, because such a power supply provides better stability of the amplifier supply voltage. On the other hand, there is nothing strange - the “pulse generator” produces interference when switching the power transistors of the main converter (and the APFC unit), which is expressed in high-frequency “bursts” on the power and ground circuits. Most often, the power supply converter operates at a frequency of 40-80 kHz, which is higher than the audio range, and therefore should not interfere with the device, but the interference spreads throughout the amplifier and disrupts the operating point of the amplification stages, which leads to intermodulation distortion, the sound becomes “harsher” . In a computer power supply, the 12 V and 5 V buses look like this:
So, the problem is not far-fetched and some effort should be spent on combating its negative manifestations.
FSP ATX-300GTF
Nothing unusual, a classic layout, except that the PFC choke introduces some element of disharmony into the picture. By the way, measuring the characteristics and magnitude of output ripple showed that the presence of this choke only leads to the fact that the power supply becomes heavier and “buzzes” a little at a load power of 250-300 W.
Removing unnecessary
A computer power supply must generate a lot of high-power voltages - 12 V, 5 V, 3.3 V, -5 V, the meaning of which is immediately lost as soon as we talk about an amplifier. In addition, the power supply contains a standby 5 V source, but it is better not to touch it and keep it unchanged - firstly, it is used to operate the main converter, and secondly, it will be possible to turn the amplifier on and off from external control or simply when sound signal at the amplifier input. This function will require the manufacture of a highly sensitive detector powered by 5 volts, and it is unlikely that anyone will make this element at the initial stage of amplifier assembly, but at least this possibility will remain. Let it be, it’s “free”.
After removing all circuits for generating output voltages, the following was obtained:
It turned out there wasn't much space, so the revision shouldn't contain too many details - it just won't fit. Wow, the requirements also included the presence of two output channels.
Choosing a method for obtaining increased output voltage
The computer power supply generates two main outputs: 12 V and 5 V, which explains the presence of only two pairs of secondary windings. How can you get a voltage higher than what was specified when designing the power supply?
1. Rewind the transformer.
2. Install a multiplier.
3. Add a second transformer.
Transformer rewinding
The first option is clear and simple in technical terms. One “but”, the design of a pulse transformer is not as simple as it might seem at first glance. There are a lot of requirements and restrictions, without fulfilling which you can get either an “extremely mediocre option” or, what is much worse, poor-quality insulation up to the point of electric shock. In a transformer, the primary winding is made of two parts. The first is located at the very beginning, and therefore does not interfere with rewinding, but the second is wound last.
The difficulties are multiplied by the fact that there is an electrostatic shield made of copper tape between the primary and secondary windings. To rewind, you will have to carefully wind the upper part of the primary winding, remove the screen and secondary windings. Then wind new secondary windings, restore the screen and primary winding. Naturally, there must be reliable insulation between the windings and the screen. The matter is aggravated by the fact that the transformer is impregnated with varnish, and therefore disassembling and reassembling it is a “fascinating” task and the quality of the modification will not be very good. However, if you have straight arms and want to try, here are some recommendations:
- The number of turns of the 12 V winding is almost always constant (seven turns), which is determined not by the parameters of the transformer, but by a single integer ratio of the number of turns of the 12 V and 5 V windings (four and three). If there are 12.6 volts per seven turns, then the “needed” voltage has 7 * (“needed”/12.6) number of turns, rounded to the nearest integer.
- When removing the 12 V and 5 V windings, count the space they occupied - the new winding should fit into the same dimensions.
- If there is space, it is better to use a wire with a diameter of 0.8-0.9 mm. If the cross-section of one wire is not enough, then it is worth increasing the number of wires, and not their cross-section (diameter)
- Wind the shielding turn of the tape extremely carefully (do not close the beginning with the end) and the insulation under and above it - the main defect of homemade transformers is the breakdown of the insulation or short-circuiting of the shielding winding. Copper tape is rigid with a sharp edge and cuts insulation easily. At home, it is better to use aluminum foil - it is much softer and there is less chance of cutting the insulation. Plus it's easier to find. Unfortunately, this approach has a small drawback - it is more difficult to connect a tap to aluminum foil.
Still, I would not recommend this conversion option for those who do not have experience in winding pulse transformers. It's not worth it, it might come out sideways. By the way, if a person understands the issue, then it is easier for him to wind the transformer completely from scratch, at least this “varnish” will not get underfoot, and the number of turns in all windings can be chosen optimally.
Multiplier
The second option is quite difficult to implement and has a number of serious disadvantages. An example of such a construction is shown in the figure:
- TV1 is a regular power supply transformer, without any modifications.
- TV1.1 – primary winding.
- TV1.3 and TV1.4 – 5 V channel windings.
- TV1.2 and TV1.5 are windings that, together with TV1.3 and TV1.4, form a 12 V channel.
What is important for the analysis is the fact that the shape of the voltage pulses at the output of the transformer is smooth top, and not “sine”, “saw” or other variations. The device works as follows: rectangular voltage pulses with a certain duty cycle follow on the primary winding. The pulse voltage on the primary winding is half the supply voltage or about 140 V at the rated mains voltage. On the secondary side, the shape of the pulses is preserved, and the amplitude depends on the number of turns and is distributed approximately as 9 V on the windings of the “5 V channel” (TV1.3 and TV1.4) and 21 V on the “12 V channel” (TV1.2 + TV1 .3 and TV1.4+ TV1.5).
Let us assume that at the moment a pulse of positive polarity is received and “+” follows at the upper terminals of the windings. Let's place the voltages at the control points:
- A = +21 V.
- B = +9 V.
- C = -9 V.
- D = -21 V.
From here you can immediately calculate the voltage in current “F”; it will be slightly less than circuit “B” by the amount of voltage drop across diode D1.
- F = +8.4 V.
At this polarity, diode D2 is closed, so the voltage at point “E” will be determined with the opposite polarity of the pulse.
- Voltage on capacitor C2 = +8.4 – (-21) = 29.4 V.
Let's change the polarity of the pulse, the voltages at the control points will change sign:
- A = -21 V.
- B = -9 V.
- C = +9 V.
- D = +21 V.
The polarity has changed and diode D2 opens. The voltage at point “F” will become slightly less than circuit “B” or about +8.4 V.
- E = +8.4 V.
- Voltage on capacitor C1 = +8.4 – (-21) = 29.4 V.
The circuit is symmetrical, so the voltages of the capacitors must be the same. From the analysis of the previous pulse polarity it follows that
- The voltage at point “F” is shifted relative to point “D” by the voltage of capacitor C2 (29.4 V) and is equal to +21 + 29.4 = +50.4 V.
There is no point in analyzing the similar state of point “E” when changing the polarity of the pulse, the circuit is symmetrical and there will be the same amount as now at point “F”, +50.4 V.
As a result, only “E” and “F” may be of interest, because they produce the output voltage. Let's collect the values at these points into a table. However, I forgot one more state, “pause” of the pulse from PWM control. This case is very simple, there is zero voltage on all windings and at points “E” and “F” the same voltage of +29.4 V is obtained, stored in the capacitors. (The analysis did not take into account the finite capacitance of the capacitors and the non-rectangular shape of the pulses).
Rectifier assembly D3 “selects” the highest voltage from the two inputs (“E” and “F”). This means that at the input of inductor L6 there will be pulses with an amplitude of 50 V with a pause of 8 V. With a PWM duty cycle of 70%, a voltage of approximately 37 volts will be generated at the output.
All of the above related to obtaining increased voltage of positive polarity. If it is necessary to generate a negative output, then the circuit should be “doubled” - add capacitors C1, C2 and C3, diodes D1 and D2, a pair of diodes in the assembly D3 and wind the second winding on the output inductor. Don't forget to change the polarity of capacitors and diodes.
This solution has only one advantage - you don’t have to do anything with the transformer. However, there is one more thing - insignificant, the voltage deviation at the output choke is of small amplitude, so the size of the choke and its inductance can be reduced. In fact, you can use the old 12V channel winding.
There are more disadvantages and they are serious:
- All pulse current flows through step-up capacitors C1 and C2.
- A very large charging current of the capacitors at the initial moment of time. In addition to reducing the life of the capacitors, high current levels can trigger the general protection of the power supply and it will turn off.
- Low output voltage regulation range.
- It is impossible to obtain more than one channel with output voltage stabilization. The outputs “+37 V” and “-37 V” are obtained according to the above diagram, but the usual “+/-12 V” will have to be generated using separate choke at an increased level of ripple with the mains frequency and low stability.
The main disadvantage of the circuit design is all current flows through capacitors C1 and C2. It is quite easy to find capacitors with suitable capacitance or ESR, but their pulse current will be low. In order not to be unfounded, we will select a suitable capacitor for the amplifier power supply in question (the output voltage corresponds to the specified conditions, the current value is up to 10 A).
Previously, I referred to general-purpose capacitors from the Jamicon series, let's see what is in this design - 2200 uF 50 V. Maximum current 2 amperes. Completely unsuitable, the capacitor will fail after a week of operation of the amplifier. Let's move on to the serious series, "Low ESR". For example, series:
| Denomination | Diameter, mm | Height, mm | ESR, mOhm | Max. current, A |
| 2200 µF 35 V | 16 (18) | 32 (25) | 40 | 3.8 (3.5) |
| 1500 µF 50 V | 16 (18) | 36 (32) | 51 | 4 (3.9) |
| 1000 µF 35 V | 13 (18) | 25 (15) | 70 | 2.5 (2.1) |
| 1000 µF 50 V | 13 (18) | 40 (20) | 70 | 3.4 (2.8) |
| 680 µF 35 V | 10 (16) | 28 (15) | 103 (86) | 2 (1.7) |
| 680 µF 50 V | 13 (16) | 30 (20) | 86 | 2.6 (2.3) |
The characteristics of an alternative design of the capacitor housing are indicated in parentheses.
I would like to note an interesting point: for the “680 µF 35 V” capacitor, the first version, in comparison with the second, carries less internal resistance and maximum current, usually the opposite happens - a decrease in ESR increases the current value. Apparently, the reason is the different surface area of the case.
If you look at ESR, then all capacitors are quite satisfactory. Well, how much can “fall” at a resistance of 40-90 mOhm at a current of 3-8 amperes? Trifle. The power supply will work. This is how “Chinese” crafts appear. By the way, a lot of high-quality products are produced in China, it is the local marketers who buy rubbish, hence the distrust in Chinese products... and in vain.
Well, okay, we’re collecting for ourselves, so we won’t do anything bad. The capacitor must withstand a current of at least 10/2 = 5 A in long-term mode, and it will not be possible to obtain such a characteristic with one capacitor. There remains the option of installing a pair or three capacitors in parallel. Two “1000 uF 35 V” capacitors will provide a current of up to 5 (4.2) amperes, which is not enough. You can take capacitors of the same rating, but a slightly higher voltage “1000 µF 50 V”, the maximum current will be 6.4 (5.6) amperes.
Taking into account the final inductance of the output choke, this option may work, but not particularly well. Let's move on to tripling the capacitors, “680 uF 35 V” will provide a current of up to 6 (5.1) A, or “680 uF 50 V” 7.8 (6.9) A. The latter option looks more fun, the power supply will be able to work for quite a long time.
As a result, it turns out that you will have to install 3*2*2=12 “680 uF 50 V” capacitors into the power supply; the result will not be the most compact device, and the space in the power supply is limited.
The circuit was simulated, but practically not tested, since I am not in the mood for such solutions. This modification option is provided at your own risk.
This section offers some options for implementing PP power supplies for amplifiers. A power supply circuit with separation of a bank of capacitors by resistors with a resistance in the range of 0.15-0.47 Ohms was proposed by L. Zuev:
Layout of the ULF power supply board by Vladimir Lepekhin in lay format
For ULF Natalie, boards were laid out for electrolytic capacitors with a landing diameter of d=30, 35 and 40 mm with snap-in terminals
Circuit with stabilized power supply for UN-a and operational amplifier on m/s M5230L
For the project, an ASR amplifier on a MOSFET with a current OOOS from Maxim_A (Andrey Konstantinovich), V. Lepekhin laid out boards for a low-power power supply unit for the amplifier and a powerful power supply unit for the output stage.
PSU board low-power top
PSU board low-power bottom
ULF top power supply board
ULF power supply board bottom
For the implementation of dual mono, power supplies will be used on the following PCBs:
BP ULF V2012EA
This power supply is used to power the VC (output stage). The board can be used to install electrolytes with Snap-in mounts with a diameter of up to 30 mm; mounting for diodes in TO220-3 and TO220-2 packages is provided, which expands the range of diodes used. PP dimensions 66 x 88 mm.
To power the UN with separate power supply, the following power supply board will be used:
BP ULF V2012EA
PP dimensions 66 x 52 mm. The diodes have a universal fit; they can also be installed in the TO220-2 housing; they fit electrolytes with a diameter of up to 25 mm.
Schematic diagram of a network switching power supply for ULF, output voltage +-25V at a current of up to 4.5A (approximately 200W). The circuit is assembled on an IR2153 chip and IRF740 transistors. Useful tips on assembling and setting up the device are provided.
I would like to offer a short overview of this scheme. Once there was a need for a person to assemble a simple ULF, and a housing from an old “radio engineering” preamplifier was found.
There is a lot of space in the case, but it was not possible to fit the network transformer; the case turned out to be too small in height. It was decided to assemble a switching power supply on the ir2153 chip, there was just one lying around idle.
Schematic diagram
Initially, the circuit with was taken as a basis - I strongly recommend not to assemble it as suggested there, otherwise you can cause a fire or explosion, a circuit with a fatal error and more than one.
Rice. 1. Switching power supply circuit, taken as a basis.
Rice. 2. Scheme of a switching power supply for UMZCH with a power of up to 200 W.
In the first circuit, the main mistake is that there is no separating capacitor between the field-effect transistors and the transformer; without this capacitor, the transistors will immediately explode when turned on, or after a couple of minutes they will become hot...
For the IR2153 microcircuit, the first pin is a power plus, since the voltage at pin 1 of the microcircuit is within 16-18 volts, the capacitor should be an order of magnitude higher in voltage, and not end-to-end as indicated in the original diagram - at 16V. You can set the capacitor to a voltage of 25V, I set it to 35V.
Let's go further, it is impossible to power the microcircuit as indicated on the original diagram through a diode and an 18K resistor!! Look how my IR2153 microcircuit is powered (Figure 2), and not directly from a 220-volt alternator (Figure 1).
In the circuit in Figure 1, a voltage surge in the network will immediately lead to the combustion of the microcircuit, it’s good if everything just stops working, otherwise the transistors will explode again.
These three errors in the diagram from Figure 1 can lead to very sad consequences!
Details and design
The filter choke for the 220 Volt power supply (Dr1) was taken from a switching power supply from a TV, any one will do, taking into account the amount of power you want to receive... Varistor - any 10 ohm, but not from a phone charger and similar low-power switching power supplies.
The inductance of 25 Volts (L) was taken from a 450-watt computer power supply, the extra windings were wound up - we leave only those wound with a thick wire.
The high-frequency transformer Tr1 was taken from the same place; I will dwell in detail on its winding from scratch. It is quite difficult to disassemble such a transformer without splitting the ferrite. To simplify the task, you need to put it on the stove and heat it up to a hundred degrees, in other words, as soon as a drop of water on the ferrite boils, you can disassemble it.
With this heating, the glue becomes soft and the ferrite halves are easily pulled out of the frame with the winding. When winding transformers in pulse circuits, it is recommended to wind the windings with several wires - up to 8 pieces at a time.
It is not at all necessary to do this; I wound the primary winding with one enameled copper wire with a diameter of 0.45 mm - 49 turns. Secondary windings II and III were wound with two wires with a diameter of 0.8 mm - 8 turns each.
We use high-speed rectifier diodes - KD213 or KD212 are suitable from domestic ones. For the latter, the load current according to the reference book is 1A, and for KD213 it is 10A. Diodes with a limit operating frequency of 100 kHz are suitable.
Instead of the IRF740 transistor, you can use the IRF840 and the like. The radiator for the transistors can be installed in half the size; under full long-term load, the transistors do not heat up very much - 45 degrees to the touch. Transistors must be placed on the radiator through insulating gaskets.
Instead of RL205 diodes, you can install any diode bridge with a maximum constant reverse voltage of 600V and a maximum constant forward current of 6A.
The transition capacitance (0.1 µF) between the transistors and the transformer must be for a voltage of 630V!
With the specified ratings, this circuit provides an output power of approximately 200 W at a current of up to 4.5A.
I didn’t make a signet for the power supply circuit - I immediately drew it on the PCB. Each part and their location options may be different. The diagram is simple and drawing your own signet will not be difficult.
Here's what I got:
Rice. 3. Plan of my printed circuit board for a switching network power supply.
As you can see from the sketch, instead of a separating capacitor between the transistors and the transformer, I have three installed. I had to do this because there wasn’t one for the required voltage, so I ended up collecting it from different capacitors with a total capacitance of 0.5 µF.
The most ideal option would be 1 µF at 630V. But everything works quite normally with both a 0.1 µF capacitance and a 0.5 µF capacitance.
Rice. 4. Finished printed circuit board for a switching power supply (view from the connections side).
Rice. 5. Finished switching power supply board (view from the parts side).
Rice. 6. Homemade network switching power supply for UMZCH.
Rice. 7. Appearance of a network switching power supply for a low-frequency power amplifier.
Setting up
After assembling the circuit, the first connection is made through a 220V 60W light bulb connected in series with the power supply.
If no errors or short circuits were made during assembly, then when turned on, the light should flash briefly and go out - this indicates that everything was assembled correctly and there is no short circuit in the circuit.
You can turn on the lamp at a suitable voltage on the low side as a load and let the circuit work for about five minutes. If nothing smokes, then you can remove the lamp at 220 and use a ready-made power supply.
If the lamp connected to the 220V power supply is lit when first turned on and does not go out, then there is a malfunction in the circuit.
Rice. 8. The switching power supply is installed in a housing with a low-frequency amplifier.
Rice. 9. ULF board and power supply for it in the housing from the Radiotekhnika preamplifier (front view).
Rice. 10. ULF board and power supply for it in the housing from the Radiotekhnika preamplifier (rear view).
As an addition: the ULF circuit is taken from.
Rice. 11. ULF circuit with an output power of 60W at a 4 Ohm load and +-28V power supply.
Literature:
- radiostroi.ru/pitan776/57-impblokpitkomp
- A. Ageev - Amplifier block of an amateur radio complex. Radio Magazine 1982, number 8.
