Refinement of powerful Chinese power supplies. Refinement of cheap Chinese ATX power supplies. it can hardly be called a stabilizer

Greetings to all readers. We have long been asked to test this switching power supply, which has become very popular among DIYers. This is a fairly cheap unit that can be used as a power source in a homemade soldering station, laboratory power supply, etc., in general, a universal thing.

The Chinese produce several versions, the circuit design is almost the same, the only difference is in the output voltage and current, my sample is 24 Volts, with a stated current of 4A and 6A if an additional cooler is used.
The board is quite compact, the overall dimensions with a small error you now see on your screens.

About the scheme. This is a single-cycle network step-down switching power supply with output voltage stabilization and current protection. The circuit was built on the basis of the not very popular CR6842 PWM controller (analogous to SG6842), for me, on the UC38XX family of microcircuits the unit would be more repairable, the original microcircuit is quite expensive.

The board is double-sided, the components are sealed well.

An example power supply circuit is shown below.

The power input is made in an interesting way, essentially these are clamps where the network wires are inserted; there is no need to solder or screw in anything.

Next comes the fuse and surge protector, everything is as it should be.

The diode bridge is a ready-made assembly KBP307 (3A, 700V).

After the bridge we see a thermistor, its initial resistance is 5 Ohms at a maximum current of 3A, designed to reduce the starting current when the unit is connected to a 220 Volt network.

A smoothing electrolyte with a capacity of 82 μF, taking into account 1 μF per 1 watt of power, everything is as it should be.

Then everything is clear - a generator microcircuit, a power N-channel field switch, in this version there is a P20NK60 transistor, judging by the marking 20 Amperes 600 Volts, it has a colossal current reserve, installed on a small radiator.

Pulses are supplied to the gate of the field-effect transistor through a limiting resistor and a diode, which is switched in the reverse direction and is designed to quickly discharge the gate capacitance of the field-effect transistor.

In the output part there is a half-wave rectifier based on a dual Schottky diode in a TO-220 package, moreover, both diodes are connected in parallel, which significantly reduces the transition resistance, and therefore heating.

After the rectifier, there is a filter, which consists of two electrolytes and a choke, moreover, one electrolyte is placed before the choke, the second after.

Well, there is an LED with a limiting resistor, which indicates the presence of output voltage.

The output voltage is controlled by an optocoupler, and the voltage is set by an adjustable zener diode TL431. By changing the ratio of the resistances of the resistive divider in the zener diode circuit, you can change the output voltage of the power supply within small limits.

In general, everything suggests that the power source is good, but we’ll still check it.
The first test is to check the output voltage.

Everything is fine, and the no-load current is only 12-13mA! which is a very good indicator.

The declared output current is 4A.

According to Uncle Ohm's law, to remove 4 amperes of current from a 24 Volt source, we need a load with a resistance of about 6 Ohms, you can use a nichrome spiral, but I had a 20-watt 5.6 Ohm resistor lying next to me, so I connected it.

The source is connected via a network wattmeter, and a low-voltage Volt/Ampere/Watt meter is used as a meter at the output.

At a current of 4.2A, the output voltage drops slightly.

With this situation, the unit consumes about 110 watts from a 220 Volt network, and the output is about 100 watts, the efficiency is around 90%, which is very good.

I tried to remove the current at 5.5A, everything was also fine, when I tried to remove more, the protection was triggered.

By the way! The protection is implemented on the hiccup principle and works well.
When a short circuit occurs, a voltage drop occurs across the current sensor, which is a low-resistance resistor connected to the source circuit of the field switch. The microcircuit monitors the fall and if the value is too high, it goes into protection.

I also took some measurements of the output voltage ripple.

Idling, division 20mV

Current 0.6A, division 20mV

Current 3.6 A division 20 mV

Current 4.2A division 20mV

The results were amazing, I thought there would be more pulsations.

At the end I left the unit working for 10 minutes, output current 3.6A

After 10 minutes, without turning off the unit, I took temperature measurements


1) On the radiator of the diode rectifier

2) On the radiator of the field key

3) Transformer windings

4) Transformer core

5) On the input diode rectifier

Advantages.

1) Compact, lightweight, well made.
2) Price, well, so-so, not too cheap and not expensive
3) Versatile
4) Excellent stabilization
5) Availability of short-circuit protection, fulfills its purpose
6) The presence of a filter both at the input and at the output; in general, the circuit is well organized.

Flaws

1) It is better to change the radiators or screw on the cooler; during long-term operation at high currents they get very hot.

2) A small-sized transformer with seemingly no power reserve, so it will overheat at high currents.

Results.

Everyone knows that the Chinese save on everything, and this power source is no exception. But taking into account its advantages, I recommend it, it is not afraid of short circuits, it is made well, the components are sealed neatly, there is protection, good stabilization, in general, everything that is needed for, say, implementation in a homemade soldering station or a simple laboratory power source, there are a lot of areas of application.

The product can be purchased

A detailed video of the test can be viewed below.

Sincerely - AKA KASYAN
MY YOUTUBE CHANNEL

The best option is to purchase and use a high-quality power supply. But if there is no opportunity and/or there is a desire to improve the unit you already have, then good results can be obtained by modifying a cheap (budget) power supply.

Chinese designers, as a rule, make printed circuit boards according to the criterion of maximum versatility, that is, in such a way that, depending on the number of installed elements, quality and, accordingly, price can be varied.
Therefore, if you install those parts that the manufacturer saved on and change a few other things, you will get a unit in the middle price category. Of course, it cannot be compared with expensive copies, where the topology of printed circuit boards, circuit design, and all the details were initially calculated to obtain high quality.

Whatever you do with your power supply, you do at your own peril and risk!
If you do not have sufficient qualifications, then do not read what is written here, much less do anything!
But for the average computer this is a completely acceptable option.

First of all, you need to open the power supply and estimate the size of the largest transformer; if it has a tag with the numbers 33 or higher at the beginning and has dimensions of 3x3x3 cm or more, it makes sense to tinker. Otherwise, you are unlikely to achieve an acceptable result.

In photo 1 there is a transformer of a normal power supply, in photo 2 there is a transformer from an outright Chinese.

You should also pay attention to the dimensions of the group stabilization choke. The larger the dimensions of the transformer and inductor cores, the greater the margin for saturation currents.
For a transformer, getting into saturation is fraught with a sharp drop in efficiency and the likelihood of failure of high-voltage switches, for a choke - a strong voltage dispersion in the main channels.

Rice. 1 Typical Chinese ATX power supply, no mains filter.

The most critical details in a power supply are:
.High voltage capacitors
.High voltage transistors
.High voltage rectifier diodes
.High frequency power transformer
.Low-voltage diode rectifier assemblies

Revision:
1. First, you need to replace the input electrolytic capacitors; replace them with capacitors of larger capacity that can fit on the seats. Typically, cheap units are rated at 220µF x 200V or at best 330µF x 200V. We change it to 470µF x 200V or better to 680µF x 200V. These capacitors affect the ability of the unit to withstand a short-term loss of mains voltage and the power supplied by the Power Supply.

Rice. 2 Input electrolytic capacitors and high-voltage part of the power supply, including a rectifier, half-bridge inverter, electrolytes at 200V (330µF, 85 degrees).

Next, you need to install all the chokes in the low-voltage part of the power supply and the line filter choke (the place for its installation).
The chokes can be wound yourself on a ferrite ring with a diameter of 1-1.5 cm using copper wire with varnish insulation with a cross-section of 1.0-2.0 mm, 10-15 turns. You can also take chokes from a faulty power supply. You also need to solder smoothing capacitors into the empty spaces of the low-voltage part. The capacitance of the capacitors should be chosen to be maximum, but so that it can fit in its standard place.
Usually it is enough to put 2200µF capacitors at 16V series Low ESR 105 degrees, in the +3.3V, +5V, +12V circuit.

In the rectifier modules of the secondary rectifiers, we replace all diodes with more powerful ones.
The energy consumption of computers has recently increased to a greater extent on the + 12V bus (motherboards and processors), so first of all you need to pay attention to this module.

Typical type of rectifier diodes:

1. - Diode assembly MBR3045PT (30A) - Installed in expensive power supplies;

2. - diode assembly UG18DCT (18A) - less reliable;

3. - diodes instead of assembly (5A) - the most unreliable option, subject to mandatory replacement.

Channel +5V Stby- We change the standby diode FR302 to 1N5822. We also install the missing filter choke there, and increase the first filter capacitor to 1000μF.

Channel +3.3V- we change the S10C45 assembly to 20C40 (20A/40V), to the existing capacity 2200uF/10V, add another 2200uF/16V and the missing inductor. If the +3.3V channel is implemented on a field device, then install a transistor with a power of at least 40A/50V (IRFZ48N).

Channel +5V- We change the diode assembly S16C45 to 30C40S. Instead of one electrolyte 1000uF/10V, we set 3300uF/10V + 1500uF/16V.

Channel +12V- We replace the F12C20 diode assembly with two in parallel UG18DCT (18A/200V) or F16C20 (16A/200V). Instead of one 1000uF/16V capacitor, we put 2 pieces 2200μF/16V.

Channel -12V- Instead of 470μF/16V, set it to 1000μF/16V.

So, we install 2 or 3 diode assemblies MOSPEC S30D40 (the number after D is the voltage - the more, the calmer we are) or F12C20C - 200V and similar in characteristics, 3 capacitors 2200 μF x 16 volts, 2 capacitors 470 μF x 200V. Electrolytes, install only low-impedance ones from the 105 degree series! - 105*C.

Rice. 3 Low-voltage part of the power supply. Rectifiers, electrolytic capacitors and chokes, some missing.

If the power supply radiators are made in the form of plates with cut petals, we bend these petals in different directions to maximize their efficiency.

Rice. 5 ATX power supply with modified cooling radiators.

Further refinement of the power supply boils down to the following... As is known in the power supply, the +5 volt and +12 volt channels are stabilized and controlled simultaneously. With +5 volts set, the actual voltage on channel +12 is 12.5 volts. If the computer has a heavy load on channel +5 (AMD-based system), then the voltage drops to 4.8 volts, while the voltage on channel +12 becomes equal to 13 volts. In the case of a Pentium-based system, the +12 volt channel is more heavily loaded and the opposite happens. Due to the fact that the +5 volt channel in the power supply is made of much higher quality, even a cheap unit will power an AMD-based system without any problems. Whereas the power consumption of the Pentium is much higher (especially at +12 volts) and the cheap power supply must be improved.
Excessive voltage on the 12 volt channel is very harmful for hard drives. Basically, HDD heating occurs due to increased voltage (more than 12.6 volts). In order to reduce the voltage of 13 volts, it is enough to solder a powerful diode, for example KD213, into the gap of the yellow wire powering the HDD. As a result, the voltage will decrease by 0.6 volts and will be 11.6 - 12.4V, which is quite safe for the hard drive.

As a result, by upgrading a cheap ATX power supply in this way, you can get a good power supply for a home computer, which will also heat up much less.

Everyone knows that there is such an operation as pre-sale preparation of goods. A simple but very necessary action. By analogy with it, I have long been using pre-use preparation of all purchased Chinese-made goods. There is always the possibility of modification in these products, and I note that it is really necessary, which is a consequence of the manufacturer saving on high-quality material for its individual elements or not installing them at all. Let me be suspicious and suggest that all this is not accidental, but is an integral element of the manufacturer’s policy aimed ultimately at reducing the service life of the manufactured product, which results in an increase in sales. Having decided to actively use a miniature electric massager (made in China, of course), I immediately noticed its power supply, which looks like a mobile phone charger and even has an inscription COURIER CHARGER- mobile charger. Having an OUTPUT of 5 volts and 500 mA. Without even being convinced of its serviceability, I took it apart and looked at the contents.

The electronic components installed on the board and especially the zener diode at the output indicated that this was indeed a power supply. By the way, I don’t consider the absence of a diode bridge to be a positive thing.

The connected load, in the form of two 2.5 V light bulbs in series, with a current consumption of 150 mA, detected 5.76 V at the output. The device is designed to be powered by three AA batteries - 4.5 V, I think acceptable, and 5 V from the adapter, but anything else, in this particular case, is clearly useless.

After searching for a schematic on the Internet, I chose to draw, based on a photo taken in advance, a printed circuit board with the electronic components located on it.

Adapter circuit and conversion

The image of the printed circuit board made it possible to draw the existing power supply circuit. The CHY 1711 transistor optocoupler, C945, S13001 transistors and other components did not allow me to call the circuit primitive, but with the existing ratings of some components and the absence of others, it did not suit me.

A 160 mA fuse was introduced into the new circuit, and instead of the existing rectifier, a diode bridge consisting of 4 1N4007 diodes was introduced. The value of the zener diode VD3 controlling the optocoupler has been changed from 4V6 to 3V6, which should reduce the output voltage to the desired level.

There was enough free space on the board so that it was not difficult to implement the planned changes. The newly assembled power supply had an output voltage of almost 4.5 volts.

And current output up to 300 mA inclusive.

As a result, some additional electronic components and time devoted to interesting work gave me the opportunity to have a decent power supply that I hope will serve faithfully for a long time. Babay was involved in debugging the power supply.

Hi all. Anyone who works in electronics should have a . If you are reluctant to solder or you are a beginner radio amateur, this article was written especially for you. Let's immediately talk about the characteristics of the power supply and its difference from popular varieties of power supplies based on LM317 or LM338.

Modules for power supply

We will assemble a switching power supply, but we will not solder anything, we will simply buy from the Chinese an already soldered voltage regulation module with current limitation, such a module can deliver 30 volts 5 amperes. Agree that not every analog power supply is capable of this, and what losses in the form of heat, since the transistor or microcircuit takes on the excess voltage. I’m not writing about a specific type of module and its circuit - there are all sorts of them.

Now the indication - here we won’t invent anything either, we’ll take a ready-made indication module, as with the voltage control module.

How will all this be powered from a 220 V network - read on. There are two ways here.

1. The first is to look for a ready-made transformer or wind your own.
2. The second is to take a switching power supply for the required voltage and current, or modify it to the required characteristics.

And yes, I forgot to say that you can supply the control module with a maximum of 32 volts without consequences, but 30 volts is better than 5 amperes, you need to be careful with the current too, since the control circuit tolerates 5 amperes, but no more, but it gives everything it has to That's why it burns out easily in a transformer.

PSU assembly

The assembly process itself is even more interesting. Let me tell you how I get on with the components.

• Switching power supply from a laptop 19 volts 3.5 amperes.
• Control module.
• Display module.

That's all, yes, I didn't forget to add anything, but we probably also need some old building. My Soviet car radio worked, and any other one will do the same, but I would like to separately praise the case from a PC DVD drive.

We are assembling our future power supply, before attaching the boards to the case, we need to insulate them, I provided a backing made of thick film and then all the boards can be attached with double-sided tape.

But when it came to variable resistors for adjusting voltage and limiting current, I realized that I didn’t have them, well, not that I didn’t have them at all - there wasn’t the required value, namely 10 K. But they are on the board, and I did the following: I found two burned-out variables (so it wouldn’t be a pity), took out the handles and thought of soldering them to the variables that were on the board, why were they - I unsoldered them and tinned the screw.

But nothing came of it; I was able to center it only when I did this nonsense through heat shrink. But it worked, I’m happy with it, and we’ll find out how long it will work.

If you wish, you can paint the body, I didn’t do it very well, but it’s better than just metal.

As a result, we have a very compact, lightweight laboratory power supply with short circuit protection, current limitation, and, of course, voltage regulation. And all this is done very smoothly thanks to multi-turn resistors that were soldered from the control board. The voltage adjustment turned out to be from 0.8 volts to 20. The current limit was from 20 mA to 4 A. Good luck to everyone, I was with you Kalyan.Super.Bos

Discuss the article HOMEMADE POWER SUPPLY USING READY MODULES

Many already know that I have a weakness for all kinds of power supplies, but here is a two-in-one review. This time there will be a review of a radio constructor that allows you to assemble the basis for a laboratory power supply and a variant of its real implementation.
I warn you, there will be a lot of photos and text, so stock up on coffee :)

First, I’ll explain a little what it is and why.
Almost all radio amateurs use such a thing as a laboratory power supply in their work. Whether it's complex with software control or completely simple on the LM317, it still does almost the same thing, powers different loads while working with them.
Laboratory power supplies are divided into three main types.
With pulse stabilization.
With linear stabilization
Hybrid.

The first ones include a switching controlled power supply, or simply a switching power supply with a step-down PWM converter. I have already reviewed several options for these power supplies. , .
Advantages - high power with small dimensions, excellent efficiency.
Disadvantages - RF ripple, presence of capacious capacitors at the output

The latter do not have any PWM converters on board; all regulation is carried out in a linear manner, where excess energy is simply dissipated on the control element.
Pros - Almost complete absence of ripple, no need for output capacitors (almost).
Cons - efficiency, weight, size.

The third is a combination of either the first type with the second, then the linear stabilizer is powered by a slave buck PWM converter (the voltage at the output of the PWM converter is always maintained at a level slightly higher than the output, the rest is regulated by a transistor operating in linear mode.
Or it is a linear power supply, but the transformer has several windings that switch as needed, thereby reducing losses on the control element.
This scheme has only one drawback, complexity, which is higher than that of the first two options.

Today we will talk about the second type of power supply, with a regulating element operating in linear mode. But let's look at this power supply using the example of a designer, it seems to me that this should be even more interesting. After all, in my opinion, this is a good start for a novice radio amateur to assemble one of the main devices.
Well, or as they say, the right power supply must be heavy :)

This review is more aimed at beginners; experienced comrades are unlikely to find anything useful in it.

For review, I ordered a construction kit that allows you to assemble the main part of a laboratory power supply.
The main characteristics are as follows (from those declared by the store):
Input voltage - 24 Volts AC
Output voltage adjustable - 0-30 Volts DC.
Output current adjustable - 2mA - 3A
Output voltage ripple - 0.01%
The dimensions of the printed board are 80x80mm.

A little about packaging.
The designer arrived in a regular plastic bag, wrapped in soft material.
Inside, in an antistatic zip-lock bag, were all the necessary components, including the circuit board.

Everything inside was a mess, but nothing was damaged; the printed circuit board partially protected the radio components.

I won’t list everything that is included in the kit, it’s easier to do this later during the review, I’ll just say that I had enough of everything, even some left over.

A little about the printed circuit board.
The quality is excellent, the circuit is not included in the kit, but all the ratings are marked on the board.
The board is double-sided, covered with a protective mask.

The board coating, tinning, and the quality of the PCB itself is excellent.
I was only able to tear off a patch from the seal in one place, and that was after I tried to solder a non-original part (why, we will find out later).
In my opinion, this is the best thing for a beginner radio amateur; it will be difficult to spoil it.

Before installation, I drew a diagram of this power supply.

The scheme is quite thoughtful, although not without its shortcomings, but I’ll tell you about them in the process.
Several main nodes are visible in the diagram; I separated them by color.
Green - voltage regulation and stabilization unit
Red - current regulation and stabilization unit
Purple - indicating unit for switching to current stabilization mode
Blue - reference voltage source.
Separately there are:
1. Input diode bridge and filter capacitor
2. Power control unit on transistors VT1 and VT2.
3. Protection on transistor VT3, turning off the output until the power supply to the operational amplifiers is normal
4. Fan power stabilizer, built on a 7824 chip.
5. R16, R19, C6, C7, VD3, VD4, VD5, unit for forming the negative pole of the power supply of operational amplifiers. Due to the presence of this unit, the power supply will not operate simply on direct current; it is the alternating current input from the transformer that is required.
6. C9 output capacitor, VD9, output protective diode.

First, I will describe the advantages and disadvantages of the circuit solution.
Pros -
It's nice to have a stabilizer to power the fan, but the fan needs 24 Volts.
I am very pleased with the presence of a power source of negative polarity; this greatly improves the operation of the power supply at currents and voltages close to zero.
Due to the presence of a source of negative polarity, protection was introduced into the circuit; as long as there is no voltage, the power supply output will be turned off.
The power supply contains a reference voltage source of 5.1 Volts, this made it possible not only to correctly regulate the output voltage and current (with this circuit, voltage and current are regulated from zero to maximum linearly, without “humps” and “dips” at extreme values), but also makes it possible to control external power supply, I simply change the control voltage.
The output capacitor has a very small capacitance, which allows you to safely test the LEDs; there will be no current surge until the output capacitor is discharged and the PSU enters current stabilization mode.
The output diode is necessary to protect the power supply from supplying reverse polarity voltage to its output. True, the diode is too weak, it is better to replace it with another one.

Minuses.
The current-measuring shunt has too high a resistance, because of this, when operating with a load current of 3 Amps, about 4.5 Watts of heat are generated on it. The resistor is designed for 5 Watts, but the heating is very high.
The input diode bridge is made up of 3 Ampere diodes. It is good to have at least 5 Ampere diodes, since the current through the diodes in such a circuit is equal to 1.4 of the output, so in operation the current through them can be 4.2 Amperes, and the diodes themselves are designed for 3 Amperes. The only thing that makes the situation easier is that the pairs of diodes in the bridge work alternately, but this is still not entirely correct.
The big minus is that the Chinese engineers, when selecting operational amplifiers, chose an op-amp with a maximum voltage of 36 Volts, but did not think that the circuit had a negative voltage source and the input voltage in this version was limited to 31 Volts (36-5 = 31 ). With an input of 24 Volts AC, DC will be about 32-33 Volts.
Those. The op amps will operate in extreme mode (36 is the maximum, standard 30).

I'll talk more about the pros and cons, as well as about modernization later, but now I'll move on to the actual assembly.

First, let's lay out everything that is included in the kit. This will make assembly easier, and it will simply be clearer to see what has already been installed and what remains.

I recommend starting the assembly with the lowest elements, since if you install the high ones first, then it will be inconvenient to install the low ones later.
It is also better to start by installing those components that are more of the same.
I'll start with resistors, and these will be 10 kOhm resistors.
The resistors are high quality and have an accuracy of 1%.
A few words about resistors. Resistors are color coded. Many may find this inconvenient. In fact, this is better than alphanumeric markings, since the markings are visible in any position of the resistor.
Don’t be afraid of color coding; at the initial stage you can use it, and over time you will be able to identify it without it.
To understand and conveniently work with such components, you just need to remember two things that will be useful to a novice radio amateur in life.
1. Ten basic marking colors
2. Series values, they are not very useful when working with precision resistors of the E48 and E96 series, but such resistors are much less common.
Any radio amateur with experience will list them simply from memory.
1, 1.1, 1.2, 1.3, 1.5, 1.6, 1.8, 2, 2.2, 2.4, 2.7, 3, 3.3, 3.6, 3.9, 4.3, 4.7, 5.1, 5.6, 6.2, 6.8, 7.5, 8.2, 9.1.
All other denominations are multiplied by 10, 100, etc. For example 22k, 360k, 39Ohm.
What does this information provide?
And it gives that if the resistor is of the E24 series, then, for example, a combination of colors -
Blue + green + yellow is impossible in it.
Blue - 6
Green - 5
Yellow - x10000
those. According to calculations, it comes out to 650k, but there is no such value in the E24 series, there is either 620 or 680, which means either the color was recognized incorrectly, or the color has been changed, or the resistor is not in the E24 series, but the latter is rare.

Okay, enough theory, let's move on.
Before installation, I shape the resistor leads, usually using tweezers, but some people use a small homemade device for this.
We are not in a hurry to throw away the cuttings of the leads; sometimes they can be useful for jumpers.

Having established the main quantity, I reached single resistors.
It may be more difficult here; you will have to deal with denominations more often.

I don’t solder the components right away, but simply bite them and bend the leads, and I bite them first and then bend them.
This is done very easily, the board is held in your left hand (if you are right-handed), and the component being installed is pressed at the same time.
We have side cutters in our right hand, we bite off the leads (sometimes even several components at once), and immediately bend the leads with the side edge of the side cutters.
This is all done very quickly, after a while it is already automatic.

Now we’ve reached the last small resistor, the value of the required one and what’s left are the same, that’s not bad :)

Having installed the resistors, we move on to diodes and zener diodes.
There are four small diodes here, these are the popular 4148, two zener diodes of 5.1 Volts each, so it’s very difficult to get confused.
We also use it to form conclusions.

On the board, the cathode is indicated by a stripe, just like on diodes and zener diodes.

Although the board has a protective mask, I still recommend bending the leads so that they do not fall on adjacent tracks; in the photo, the diode lead is bent away from the track.

The zener diodes on the board are also marked as 5V1.

There are not very many ceramic capacitors in the circuit, but their markings can confuse a novice radio amateur. By the way, it also obeys the E24 series.
The first two digits are the nominal value in picofarads.
The third digit is the number of zeros that must be added to the denomination
Those. for example 331 = 330pF
101 - 100pF
104 - 100000pF or 100nF or 0.1uF
224 - 220000pF or 220nF or 0.22uF

The main number of passive elements has been installed.

After that, we move on to installing operational amplifiers.
I would probably recommend buying sockets for them, but I soldered them as is.
On the board, as well as on the chip itself, the first pin is marked.
The remaining conclusions are counted counterclockwise.
The photo shows the place for the operational amplifier and how it should be installed.

For microcircuits, I do not bend all the pins, but only a couple, usually these are the outer pins diagonally.
Well, it’s better to bite them so that they stick out about 1mm above the board.

That's it, now you can move on to soldering.
I use a very ordinary soldering iron with temperature control, but a regular soldering iron with a power of about 25-30 watts is quite sufficient.
Solder 1mm in diameter with flux. I specifically do not indicate the brand of solder, since the solder on the coil is not original (original coils weigh 1Kg), and few people will be familiar with its name.

As I wrote above, the board is of high quality, soldered very easily, I did not use any fluxes, only what is in the solder is enough, you just need to remember to sometimes shake off the excess flux from the tip.

Here I took a photo with an example of good soldering and not so good one.
A good solder should look like a small droplet enveloping the terminal.
But there are a couple of places in the photo where there is clearly not enough solder. This will happen on a double-sided board with metallization (where the solder also flows into the hole), but this cannot be done on a single-sided board; over time, such soldering may “fall off”.

The terminals of the transistors also need to be pre-formed; this must be done in such a way that the terminal does not become deformed near the base of the case (elders will remember the legendary KT315, whose terminals loved to break off).
I shape powerful components a little differently. Molding is done so that the component stands above the board, in which case less heat will transfer to the board and will not destroy it.

This is what molded powerful resistors look like on a board.
All components were soldered only from below, the solder that you see on the top of the board penetrated through the hole due to capillary effect. It is advisable to solder so that the solder penetrates a little to the top, this will increase the reliability of the soldering, and in the case of heavy components, their better stability.

If before this I molded the terminals of the components using tweezers, then for the diodes you will already need small pliers with narrow jaws.
The conclusions are formed in approximately the same way as for resistors.

But there are differences during installation.
If for components with thin leads installation occurs first, then biting occurs, then for diodes the opposite is true. You simply won’t bend such a lead after biting it, so first we bend the lead, then bite off the excess.

The power unit is assembled using two transistors connected according to a Darlington circuit.
One of the transistors is installed on a small radiator, preferably through thermal paste.
The kit included four M3 screws, one goes here.

A couple of photos of the nearly soldered board. I will not describe the installation of terminal blocks and other components; it is intuitive and can be seen from the photograph.
By the way, about the terminal blocks, the board has terminal blocks for connecting the input, output, and fan power.

I haven’t washed the board yet, although I often do it at this stage.
This is due to the fact that there will still be a small part to finalize.

After the main assembly stage we are left with the following components.
Powerful transistor
Two variable resistors
Two connectors for board installation
Two connectors with wires, by the way the wires are very soft, but of small cross-section.
Three screws.

Initially, the manufacturer intended to place variable resistors on the board itself, but they are placed so inconveniently that I didn’t even bother to solder them and showed them just as an example.
They are very close and it will be extremely inconvenient to adjust, although it is possible.

But thank you for not forgetting to include the wires with connectors, it’s much more convenient.
In this form, the resistors can be placed on the front panel of the device, and the board can be installed in a convenient place.
At the same time, I soldered a powerful transistor. This is an ordinary bipolar transistor, but it has a maximum power dissipation of up to 100 Watts (naturally, when installed on a radiator).
There are three screws left, I don’t even understand where to use them, if in the corners of the board, then four are needed, if you are attaching a powerful transistor, then they are short, in general it’s a mystery.

The board can be powered from any transformer with an output voltage of up to 22 Volts (the specifications state 24, but I explained above why such a voltage cannot be used).
I decided to use a transformer that had been lying around for a long time for the Romantic amplifier. Why for, and not from, and because it hasn’t stood anywhere yet :)
This transformer has two output power windings of 21 Volts, two auxiliary windings of 16 Volts and a shield winding.
The voltage is indicated for the input 220, but since we now already have a standard of 230, the output voltages will be slightly higher.
The calculated power of the transformer is about 100 watts.
I paralleled the output power windings to get more current. Of course, it was possible to use a rectification circuit with two diodes, but it would not work better, so I left it as is.

For those who don’t know how to determine the power of a transformer, I made a short video.

First trial run. I installed a small heatsink on the transistor, but even in this form there was quite a lot of heating, since the power supply is linear.
Adjustment of current and voltage occurs without problems, everything worked right away, so I can already fully recommend this designer.
The first photo is voltage stabilization, the second is current.

First, I checked what the transformer outputs after rectification, as this determines the maximum output voltage.
I got about 25 Volts, not a lot. The capacity of the filter capacitor is 3300 μF, I would advise increasing it, but even in this form the device is quite functional.

Since for further testing it was necessary to use a normal radiator, I moved on to assembling the entire future structure, since the installation of the radiator depended on the intended design.
I decided to use the Igloo7200 radiator I had lying around. According to the manufacturer, such a radiator is capable of dissipating up to 90 watts of heat.

The device will use a Z2A housing based on a Polish-made idea, the price will be about $3.

Initially, I wanted to move away from the case that my readers are tired of, in which I collect all sorts of electronic things.
To do this, I chose a slightly smaller case and bought a fan with a mesh for it, but I couldn’t fit all the stuffing into it, so I purchased a second case and, accordingly, a second fan.
In both cases I bought Sunon fans, I really like the products of this company, and in both cases I bought 24 Volt fans.

This is how I planned to install the radiator, board and transformer. There is even a little room left for the filling to expand.
There was no way to get the fan inside, so it was decided to place it outside.

We mark the mounting holes, cut the threads, and screw them for fitting.

Since the selected case has an internal height of 80mm, and the board also has this size, I secured the radiator so that the board is symmetrical with respect to the radiator.

The leads of the powerful transistor also need to be slightly molded so that they do not become deformed when the transistor is pressed against the radiator.

A small digression.
For some reason, the manufacturer thought of a place to install a rather small radiator, because of this, when installing a normal one, it turns out that the fan power stabilizer and the connector for connecting it get in the way.
I had to unsolder them, and seal the place where they were with tape so that there would be no connection to the radiator, since there is voltage on it.

I cut off the excess tape on the back side, otherwise it would turn out completely sloppy, we’ll do it according to Feng Shui :)

This is what a printed circuit board looks like with the heatsink finally installed, the transistor is installed using thermal paste, and it is better to use good thermal paste, since the transistor dissipates power comparable to a powerful processor, i.e. about 90 watts.
At the same time, I immediately made a hole for installing the fan speed controller board, which in the end still had to be re-drilled :)

To set zero, I unscrewed both knobs to the extreme left position, turned off the load and set the output to zero. Now the output voltage will be regulated from zero.

Next are some tests.
I checked the accuracy of maintaining the output voltage.
Idling, voltage 10.00 Volts
1. Load current 1 Ampere, voltage 10.00 Volts
2. Load current 2 Amps, voltage 9.99 Volts
3. Load current 3 Amperes, voltage 9.98 Volts.
4. Load current 3.97 Amperes, voltage 9.97 Volts.
The characteristics are quite good, if desired, they can be improved a little more by changing the connection point of the voltage feedback resistors, but as for me, it’s enough as is.

I also checked the ripple level, the test took place at a current of 3 Amps and an output voltage of 10 Volts

The ripple level was about 15mV, which is very good, but I thought that in fact the ripples shown in the screenshot were more likely to come from the electronic load than from the power supply itself.

After that, I started assembling the device itself as a whole.
I started by installing the radiator with the power supply board.
To do this, I marked the installation location of the fan and the power connector.
The hole was marked not quite round, with small “cuts” at the top and bottom, they are needed to increase the strength of the back panel after cutting the hole.
The biggest difficulty is usually holes of complex shape, for example, for a power connector.

A big hole is cut out of a big pile of small ones :)
A drill + a 1mm drill bit sometimes works wonders.
We drill holes, lots of holes. It may seem long and tedious. No, on the contrary, it is very fast, completely drilling a panel takes about 3 minutes.

After that, I usually set the drill a little larger, for example 1.2-1.3mm, and go through it like a cutter, I get a cut like this:

After this, we take a small knife in our hands and clean out the resulting holes, at the same time we trim the plastic a little if the hole is a little smaller. The plastic is quite soft, making it comfortable to work with.

The last stage of preparation is to drill the mounting holes; we can say that the main work on the back panel is finished.

We install the radiator with the board and the fan, try on the resulting result, and if necessary, “finish it with a file.”

Almost at the very beginning I mentioned revision.
I'll work on it a little.
To begin with, I decided to replace the original diodes in the input diode bridge with Schottky diodes; for this I bought four 31DQ06 pieces. and then I repeated the mistake of the board developers, by inertia buying diodes for the same current, but it was necessary for a higher one. But still, the heating of the diodes will be less, since the drop on Schottky diodes is less than on conventional ones.
Secondly, I decided to replace the shunt. I was not satisfied not only with the fact that it heats up like an iron, but also with the fact that it drops about 1.5 Volts, which can be used (in the sense of a load). To do this, I took two domestic 0.27 Ohm 1% resistors (this will also improve stability). Why the developers didn’t do this is unclear; the price of the solution is absolutely the same as in the version with a native 0.47 Ohm resistor.
Well, rather as an addition, I decided to replace the original 3300 µF filter capacitor with a higher quality and capacious Capxon 10000 µF...

This is what the resulting design looks like with replaced components and an installed fan thermal control board.
It turned out a little collective farm, and besides, I accidentally tore off one spot on the board when installing powerful resistors. In general, it was possible to safely use less powerful resistors, for example one 2-Watt resistor, I just didn’t have one in stock.

A few components were also added to the bottom.
A 3.9k resistor, parallel to the outermost contacts of the connector for connecting a current control resistor. It is needed to reduce the regulation voltage since the voltage on the shunt is now different.
A pair of 0.22 µF capacitors, one in parallel with the output from the current control resistor, to reduce interference, the second is simply at the output of the power supply, it is not particularly needed, I just accidentally took out a pair at once and decided to use both.

The entire power section is connected, and a board with a diode bridge and a capacitor for powering the voltage indicator is installed on the transformer.
By and large, this board is optional in the current version, but I couldn’t raise my hand to power the indicator from the maximum 30 Volts for it and I decided to use an additional 16 Volt winding.

The following components were used to organize the front panel:
Load connection terminals
Pair of metal handles
Power switch
Red filter, declared as a filter for KM35 housings
To indicate current and voltage, I decided to use the board I had left over after writing one of the reviews. But I was not satisfied with the small indicators and therefore larger ones with a digit height of 14mm were purchased, and a printed circuit board was made for them.

In general, this solution is temporary, but I wanted to do it carefully even temporarily.

Several stages of preparing the front panel.
1. Draw a full-size layout of the front panel (I use the usual Sprint Layout). The advantage of using identical housings is that preparing a new panel is very simple, since the required dimensions are already known.
We attach the printout to the front panel and drill marking holes with a diameter of 1 mm in the corners of the square/rectangular holes. Use the same drill to drill the centers of the remaining holes.
2. Using the resulting holes, we mark the cutting locations. We change the tool to a thin disk cutter.
3. We cut straight lines, clearly in size at the front, a little larger at the back, so that the cut is as complete as possible.
4. Break out the cut pieces of plastic. I usually don't throw them away because they can still be useful.

In the same way as preparing the back panel, we process the resulting holes using a knife.
I recommend drilling large-diameter holes; it does not “bite” the plastic.

We try on what we got and, if necessary, modify it using a needle file.
I had to slightly widen the hole for the switch.

As I wrote above, for the display I decided to use the board left over from one of the previous reviews. In general, this is a very bad solution, but for a temporary option it is more than suitable, I will explain why later.
We unsolder the indicators and connectors from the board, call the old indicators and the new ones.
I wrote out the pinout of both indicators so as not to get confused.
In the native version, four-digit indicators were used, I used three-digit ones. since it didn’t fit into my window anymore. But since the fourth digit is needed only to display the letter A or U, their loss is not critical.
I placed the LED indicating the current limit mode between the indicators.

I prepare everything necessary, solder a 50 mOhm resistor from the old board, which will be used as before, as a current-measuring shunt.
This is the problem with this shunt. The fact is that in this option I will have a voltage drop at the output of 50 mV for every 1 Ampere of load current.
There are two ways to get rid of this problem: use two separate meters, for current and voltage, while powering the voltmeter from a separate power source.
The second way is to install a shunt in the positive pole of the power supply. Both options did not suit me as a temporary solution, so I decided to step on the throat of my perfectionism and make a simplified version, but far from the best.

For the design, I used mounting posts left over from the DC-DC converter board.
With them I got a very convenient design: the indicator board is attached to the ampere-voltmeter board, which in turn is attached to the power terminal board.
It turned out even better than I expected :)
I also placed a current-measuring shunt on the power terminal board.

The resulting front panel design.

And then I remembered that I forgot to install a more powerful protective diode. I had to solder it later. I used a diode left over from replacing the diodes in the input bridge of the board.
Of course, it would be nice to add a fuse, but this is no longer in this version.

But I decided to install better current and voltage control resistors than those suggested by the manufacturer.
The original ones are quite high quality and run smoothly, but these are ordinary resistors and, in my opinion, a laboratory power supply should be able to more accurately adjust the output voltage and current.
Even when I was thinking about ordering a power supply board, I saw them in the store and ordered them for review, especially since they had the same rating.

In general, I usually use other resistors for such purposes; they combine two resistors inside themselves for rough and smooth adjustment, but lately I can’t find them on sale.
Does anyone know their imported analogues?

The resistors are of quite high quality, the rotation angle is 3600 degrees, or in simple terms - 10 full turns, which provides a change of 3 Volts or 0.3 Amperes per 1 turn.
With such resistors, the adjustment accuracy is approximately 11 times more accurate than with conventional ones.

New resistors compared to the original ones, the size is certainly impressive.
Along the way, I shortened the wires to the resistors a little, this should improve noise immunity.

I packed everything into the case, in principle there is even a little space left, there is room to grow :)

I connected the shielding winding to the grounding conductor of the connector, the additional power board is located directly on the terminals of the transformer, this is of course not very neat, but I have not yet come up with another option.

Check after assembly. Everything started almost the first time, I accidentally mixed up two digits on the indicator and for a long time I could not understand what was wrong with the adjustment, after switching everything became as it should.

The last stage is gluing the filter, installing the handles and assembling the body.
The filter has a thinner edge around its perimeter, the main part is recessed into the housing window, and the thinner part is glued with double-sided tape.
The handles were originally designed for a shaft diameter of 6.3mm (if I’m not mistaken), the new resistors have a thinner shaft, so I had to put a couple of layers of heat shrink on the shaft.
I decided not to design the front panel in any way for now, and there are two reasons for this:
1. The controls are so intuitive that there is no particular point in the inscriptions yet.
2. I plan to modify this power supply, so changes in the design of the front panel are possible.

A couple of photos of the resulting design.
Front view:

Back view.
Attentive readers will probably have noticed that the fan is positioned in such a way that it blows hot air out of the case, rather than pumping cold air between the fins of the radiator.
I decided to do this because the radiator is slightly smaller in height than the case, and to prevent hot air from getting inside, I installed the fan in reverse. This, of course, significantly reduces the efficiency of heat removal, but allows for a little ventilation of the space inside the power supply.
Additionally, I would recommend making several holes at the bottom of the lower half of the body, but this is more of an addition.

After all the alterations, I ended up with a slightly less current than in the original version, and was about 3.35 Amperes.

So, I’ll try to describe the pros and cons of this board.
pros
Excellent workmanship.
Almost correct circuit design of the device.
A complete set of parts for assembling the power supply stabilizer board
Well suited for beginner radio amateurs.
In its minimal form, it additionally requires only a transformer and a radiator; in a more advanced form, it also requires an ampere-voltmeter.
Fully functional after assembly, although with some nuances.
No capacitive capacitors at the power supply output, safe when testing LEDs, etc.

Minuses
The type of operational amplifiers is incorrectly selected, because of this the input voltage range must be limited to 22 Volts.
Not a very suitable current measurement resistor value. It operates in its normal thermal mode, but it is better to replace it, since the heating is very high and can harm surrounding components.
The input diode bridge operates at maximum, it is better to replace the diodes with more powerful ones

My opinion. During the assembly process, I got the impression that the circuit was designed by two different people, one applied the correct regulation principle, reference voltage source, negative voltage source, protection. The second one incorrectly selected the shunt, operational amplifiers and diode bridge for this purpose.
I really liked the circuit design of the device, and in the modification section, I first wanted to replace the operational amplifiers, I even bought microcircuits with a maximum operating voltage of 40 Volts, but then I changed my mind about modifications. but otherwise the solution is quite correct, the adjustment is smooth and linear. Of course there is heating, you can’t live without it. In general, as for me, this is a very good and useful constructor for a beginning radio amateur.
Surely there will be people who will write that it is easier to buy a ready-made one, but I think that assembling it yourself is both more interesting (probably this is the most important thing) and more useful. In addition, many people quite easily have at home a transformer and a radiator from an old processor, and some kind of box.

Already in the process of writing the review, I had an even stronger feeling that this review will be the beginning in a series of reviews dedicated to the linear power supply; I have thoughts on improvement -
1. Conversion of the indication and control circuit into a digital version, possibly with connection to a computer
2. Replacing operational amplifiers with high-voltage ones (I don’t know which ones yet)
3. After replacing the op-amp, I want to make two automatically switching stages and expand the output voltage range.
4. Change the principle of current measurement in the display device so that there is no voltage drop under load.
5. Add the ability to turn off the output voltage with a button.

That's probably all. Perhaps I’ll remember and add something else, but I’m most looking forward to comments with questions.
We also plan to devote several more reviews to constructors for beginner radio amateurs; perhaps someone will have suggestions regarding certain constructors.

Not for the faint of heart

At first I didn’t want to show it, but then I decided to take a photo anyway.
On the left is the power supply that I used for many years before.
This is a simple linear power supply with an output of 1-1.2 Amperes at a voltage of up to 25 Volts.
So I wanted to replace it with something more powerful and correct.

The product was provided for writing a review by the store. The review was published in accordance with clause 18 of the Site Rules.