The simplest DIY welding inverter. Do-it-yourself welding machine Do-it-yourself inverter welding machine diagrams

Inverter welding is a modern device that is widely popular due to the light weight of the device and its dimensions. The inverter mechanism is based on the use of field-effect transistors and power switches. To become the owner of a welding machine, you can visit any tool store and acquire such a useful thing. But there is a much more economical way, which is due to the creation of inverter welding with your own hands. It is the second method that we will pay attention to in this material and consider how to do welding at home, what is needed for this and what the diagrams look like.

Features of the inverter operation

An inverter-type welding machine is nothing more than a power supply, the one that is now used in modern computers. What is the operation of the inverter based on? The following picture of electrical energy conversion is observed in the inverter:

2) Current with a constant sinusoid is converted into alternating current with a high frequency.

3) The voltage value decreases.

4) The current is rectified while maintaining the required frequency.

A list of such electrical circuit transformations is necessary in order to be able to reduce the weight of the device and its overall dimensions. After all, as you know, old welding machines, the principle of which is based on reducing the voltage and increasing the current on the secondary winding of the transformer. As a result, due to the high current value, the possibility of arc welding of metals is observed. In order for the current to increase and the voltage to decrease, the number of turns on the secondary winding decreases, but the cross-section of the conductor increases. As a result, you can notice that a transformer-type welding machine not only has significant dimensions, but also a decent weight.

To solve the problem, an option was proposed for implementing a welding machine using an inverter circuit. The principle of the inverter is based on increasing the frequency of the current to 60 or even 80 kHz, thereby reducing the weight and dimensions of the device itself. All that was required to implement an inverter welding machine was to increase the frequency thousands of times, which became possible thanks to the use of field-effect transistors.

Transistors provide communication with each other at a frequency of about 60-80 kHz. The transistor power supply circuit receives a constant current value, which is ensured by the use of a rectifier. A diode bridge is used as a rectifier, and capacitors provide voltage equalization.

Alternating current that is transferred after passing through transistors to a step-down transformer. But at the same time, a coil that is hundreds of times smaller is used as a transformer. Why a coil is used, because the frequency of the current that is supplied to the transformer is already increased 1000 times thanks to field-effect transistors. As a result, we obtain similar data as with transformer welding, only with a large difference in weight and dimensions.

What is needed to assemble an inverter

To assemble inverter welding yourself, you need to know that the circuit is designed, first of all, for a consuming voltage of 220 Volts and a current of 32 Amps. After energy conversion, the output current will increase almost 8 times and reach 250 Amperes. This current is sufficient to create a strong seam with an electrode at a distance of up to 1 cm. To implement an inverter-type power supply, you will need to use the following components:

1) A transformer consisting of a ferrite core.

2) Winding of the primary transformer with 100 turns of wire with a diameter of 0.3 mm.

3) Three secondary windings:

— internal: 15 turns and wire diameter 1 mm;

- medium: 15 turns and diameter 0.2 mm;

— external: 20 turns and diameter 0.35 mm.

In addition, to assemble the transformer, you will need the following elements:

- copper wires;

- fiberglass;

— textolite;

— electrical steel;

- cotton material.

What does an inverter welding circuit look like?

In order to understand what an inverter welding machine is, it is necessary to consider the diagram presented below.

Electrical circuit of inverter welding

All these components must be combined and thereby obtain a welding machine, which will be an indispensable assistant when performing plumbing work. Below is a schematic diagram of inverter welding.

Inverter welding power supply diagram

The board on which the device's power supply is located is mounted separately from the power section. The separator between the power part and the power supply is a metal sheet connected electrically to the unit body.

To control the gates, conductors are used, which must be soldered close to the transistors. These conductors are connected to each other in pairs, and the cross-section of these conductors does not play a special role. The only thing that is important to consider is the length of the conductors, which should not exceed 15 cm.

For a person who is not familiar with the basics of electronics, reading this kind of circuit is problematic, not to mention the purpose of each element. Therefore, if you do not have skills in working with electronics, then it is better to ask a familiar specialist to help you figure it out. For example, below is a diagram of the power part of an inverter welding machine.

Diagram of the power part of inverter welding

How to assemble inverter welding: step-by-step description + (Video)

To assemble an inverter welding machine, you must complete the following work steps:

1) Frame. It is recommended to use an old computer system unit as a housing for welding. It is best suited as it has the required number of holes for ventilation. You can use an old 10-liter canister in which you can cut holes and place the cooler. To increase the strength of the structure, it is necessary to place metal corners from the system housing, which are secured using bolted connections.

2) Assembling the power supply. An important element of the power supply is the transformer. It is recommended to use 7x7 or 8x8 ferrite as the base of the transformer. For the primary winding of the transformer, it is necessary to wind the wire across the entire width of the core. This important feature entails improved operation of the device when voltage surges occur. It is imperative to use PEV-2 copper wires as wire, and if there is no busbar, the wires are connected into one bundle. Fiberglass is used to insulate the primary winding. On top, after the layer of fiberglass, it is necessary to wind turns of shielding wires.

Transformer with primary and secondary windings for creating inverter welding

3) Power section. A step-down transformer acts as a power unit. Two types of cores are used as a core for a step-down transformer: Ш20х208 2000 nm. It is important to provide a gap between both elements, which is solved by placing newsprint. The secondary winding of a transformer is characterized by winding turns in several layers. It is necessary to lay three layers of wires on the secondary winding of the transformer, and fluoroplastic gaskets are installed between them. It is important to place a reinforced insulating layer between the windings, which will avoid voltage breakdown on the secondary winding. It is necessary to install a capacitor with a voltage of at least 1000 Volts.

Transformers for the secondary winding from old TVs

To ensure air circulation between the windings, it is necessary to leave an air gap. A current transformer is assembled on a ferrite core, which is connected to the circuit to the positive line. The core must be wrapped with thermal paper, so it is best to use cash register tape as this paper. Rectifier diodes are attached to the aluminum radiator plate. The outputs of these diodes should be connected with bare wires with a cross-section of 4 mm.

3) Inverter block. The main purpose of an inverter system is to convert direct current into high-frequency alternating current. To ensure an increase in frequency, special field-effect transistors are used. After all, it is the transistors that work to open and close at high frequencies.

It is recommended to use more than one powerful transistor, but it is best to implement a circuit based on 2 less powerful ones. This is necessary in order to be able to stabilize the current frequency. The circuit cannot do without capacitors, which are connected in series and make it possible to solve the following problems:

Aluminum plate inverter

4) Cooling system. Cooling fans should be installed on the case wall, and for this you can use computer coolers. They are necessary to ensure cooling of the working elements. The more fans you use, the better. In particular, it is imperative to install two fans to blow over the secondary transformer. One cooler will blow on the radiator, thereby preventing overheating of the working elements - rectifier diodes. The diodes are mounted on the radiator as follows, as shown in the photo below.

Rectifier bridge on the cooling radiator

Photo of the thermostat

It is recommended to install it on the heating element itself. This sensor will be triggered when the critical heating temperature of the working element is reached. When it is triggered, the power to the inverter device will be turned off.

Powerful fan for cooling the inverter device

During operation, inverter welding heats up very quickly, so the presence of two powerful coolers is a prerequisite. These coolers or fans are located on the device body so that they work to extract air.

Apply Fresh air into the system thanks to the holes in the device body. The system unit already has these holes, and if you use any other material, do not forget to provide a flow of fresh air.

5) Soldering the board is a key factor since the board is what the entire circuit is based on. It is important to install diodes and transistors on the board in opposite directions to each other. The board is mounted directly between the cooling radiators, with the help of which the entire circuit of electrical appliances is connected. The supply circuit is designed for a voltage of 300 V. The additional arrangement of capacitors with a capacity of 0.15 μF makes it possible to dump excess power back into the circuit. At the output of the transformer there are capacitors and snubbers, with the help of which the overvoltages at the output of the secondary winding are suppressed.

6) Setting up and debugging work. After the inverter welding has been assembled, several more procedures will need to be carried out, in particular, setting up the operation of the unit. To do this, connect a voltage of 15 volts to the PWM (pulse width modulator) and power the cooler. Additionally connected to the relay circuit through resistor R11. The relay is included in the circuit in order to avoid voltage surges in the 220 V network. It is imperative to monitor the relay’s activation, and then apply power to the PWM. As a result, a picture should be observed in which rectangular areas in the PWM diagram should disappear.

Device homemade inverter with a description of the elements

You can judge whether the circuit is connected correctly if the relay outputs 150 mA during setup. If a weak signal is observed, this indicates that the board connection is incorrect. There may be a breakdown in one of the windings, so to eliminate interference you will need to shorten all power supply wires.

Inverter welding in a computer system case

Checking the device's functionality

After all the assembly and debugging work has been completed, all that remains is to check the functionality of the resulting welding machine. To do this, the device is powered from a 220 V electrical network, then it is set high performance current strength and the readings are verified using an oscilloscope. In the lower loop, the voltage should be within 500 V, but not more than 550 V. If everything is done correctly with a strict selection of electronics, then the voltage indicator will not exceed 350 V.

So, now you can check the welding in action, for which we use the necessary electrodes and cut the seam until the electrode burns out completely. After this, it is important to monitor the temperature of the transformer. If the transformer simply boils, then the circuit has its shortcomings and it is better not to continue the work process.

After cutting 2-3 seams, the radiators will heat up to a high temperature, so after this it is important to allow them to cool down. To do this, a 2-3 minute pause is enough, as a result of which the temperature will drop to the optimal value.

Checking the welding machine

How to use a homemade device

After connecting a homemade device to the circuit, the controller will automatically set a certain current strength. If the wire voltage is less than 100 Volts, this indicates a malfunction of the device. You will have to disassemble the device and recheck the correct assembly again.

Using this type of welding machine, you can solder not only ferrous, but also non-ferrous metals. In order to assemble a welding machine, you will need not only knowledge of the basics of electrical engineering, but also free time to implement the idea.

Inverter welding is an indispensable thing in any owner’s garage, so if you have not yet acquired such a tool, then you can make it yourself.

It is possible to make an inverter yourself, even in the absence of deep knowledge in the field of electrical engineering and electronics. To do this, you just need to understand the operating principle of such a device and strictly adhere to the finished circuit. If you start making a homemade welding machine, which will be practically equal in technical characteristics to its factory counterpart, you can save a lot of money.

There is no doubt that a welding unit made independently will work effectively. The device, assembled according to the simplest scheme, will allow you to cook with electrodes of 3.0-5.0 mm, with an arc length of 1 cm.

1. The unnecessary computer unit may be the installation housing.
2. The configuration of the DIY welding inverter is unoriginal and resembles most other home-made designs. Many elements can be replaced with analogues. If you have the basic design details, you can calculate the optimal parameters of the housing and begin its manufacture.
3. Ready-made radiators from old devices, for example, PC power supplies, are suitable. But you can make them yourself if you have an aluminum tire on hand, the thickness of which is from 2 to 4 mm and the width is more than 3 cm. You can use a fan from any old device.
4. It is recommended to initially lay out all large parts on a plane so that the connection possibilities can be clearly determined according to the diagram.
5. Next you need to decide on a place for the fan. It should not drive a hot air stream from one element of the device to another. If there are difficulties in this situation, then you can use several fans at the same time, which will work for exhaust. The price of coolers and their weight are insignificant, but the reliability of the unit as a whole will increase significantly.

https://youtu.be/mwk1co6delA

6. The main design elements of a homemade semi-automatic welding machine, characterized by its large size and weight, are a choke and a transformer. It is recommended to place them along the edges (symmetrically to each other) or in the center. That is, their mass should not pull the device to one side. For example, working with a machine suspended on a belt over the welder’s shoulder is quite inconvenient when it constantly slides in one direction.
7. After all the parts from the welding inverter are placed in their places, it is necessary to determine the parameters of the bottom for the unit, cut it from the material at hand, which must be non-conductive. Most often, fiberglass laminate, getinax, is used for these purposes. If of this material no, then ordinary wood, pre-treated with moisture-resistant, fire-resistant solutions, will do. The extreme option even has some advantages.
8. Fastening components are usually screws, which simplifies and reduces the cost of assembling the product.

Homemade welding: materials for manufacturing, main characteristics

After assembling the semi-automatic welding inverter according to the standard simple electrical diagram, you will become the owner of an effective installation with the following performance characteristics:

• voltage – 220V;
• input current – ​​32A, output – 250A.

The diagram of welding equipment with similar technical indicators includes the following parts:

• power unit;
• power block;
• power switch drivers.

Before assembling a homemade welding machine, it is recommended to prepare all the components according to the diagram and the tools for assembly. For this homemade product you will need:

• screwdriver set;
• hacksaw for metal;
• wire, strips of copper;
• soldering iron for connecting parts of electronic circuits;
• thin metal sheet:
• threaded fastener components;
• components for forming electronic circuits;
• textolite;
• thermal paper;
• mica;
• fiberglass

For home use, inverters are often manufactured that operate from a standard power supply (220V). If there is a need, you can also assemble a device that will operate from a three-phase power supply (380V). Inverters of this type have their own advantages, one of which is a fairly high efficiency, in contrast to single-phase products.

Winding the transformer

To wind the transformer you will need a strip of copper: thickness – 0.3 mm, width – 40 mm. Copper wire is suitable for high heat. The thermal layer can be made from paper used for cash registers or photocopier paper. But the second option is worse, the paper is not strong enough and can tear.

Lacquered fabric is the best available insulating material; it is advisable to use a minimum layer. For electrical safety, the devices can be placed in the windings with PCB plates. The voltage depends on the quality of the insulation between the windings. The length of the paper strips should be enough to completely cover the perimeter of the winding and there should still be a margin - at least 2 cm.

It is forbidden to use thick wire, since the operation of the inverter welding machine is based on high-frequency currents. If you take such a wire, then its core will not be used during operation. As a result, the transformer may overheat.

In order to prevent such an effect, it is recommended to use a conductor of minimal thickness and a larger area. A surface of this type will not overheat and is an effective conductor.

When making the secondary winding, it is recommended to use 3 strips of copper, separated from each other by a fluoroplastic plate. And again a thermal layer is made of paper cash register tape. The disadvantage of this paper is that it darkens after heating, but remains tensile.

Instead of a copper strip, you can also use PEV wire - diameter no more than 0.7 mm. This wire has a large number of cores - this is its main advantage. But this type of winding is much worse than copper; wires of this type have significant air gaps, which makes them difficult to connect.

When using PEV, the design of the semi-automatic inverter has four windings (PEV with a diameter of 0.3 mm is used):

• primary winding - 100 turns;
• 1st secondary winding - 15 turns;
• 2nd secondary winding – 15 turns;
• 3rd secondary winding - 20 turns.

A cooling fan for the transformer and the entire structure is required. A system unit cooler (220V, 0.15A) is perfect for these purposes.

Cooling

The power components of the circuit of a homemade welding inverter, made independently, heat up significantly. This can contribute to rapid breakdown. To prevent them from overheating, in addition to cooling radiators for the units, additional fans must be installed.

If you have a high-power fan, you can get by with just that. In this case, the flow of cold air must be directed to the power transformer. When using low-power fans, for example, from old PCs, you need about six of them, three of which will cool the transformer.


Also, in order to prevent overheating of the welding machine with your own hands, it is recommended to install a temperature sensor on the hottest radiator, which, when the maximum permissible temperature is reached, will send a signal to automatically turn off.

For efficient operation of the ventilation system in the housing of the welding unit, it is necessary to correctly install air intakes, the grilles of which should not be blocked.

Settings

A homemade welding inverter is easy to assemble and does not require significant investment. But setting it up without involving a specialist is problematic. How to make and configure a homemade inverter yourself?

Instructions

1. It is necessary to first apply voltage to the welding unit board. The block will begin to emit a characteristic squeak. Mains voltage also needs to be supplied to the cooling fan, which will prevent parts from overheating and the unit will operate more stable.
2. When the power capacitors have received sufficient charge, it is necessary to close the current-limiting resistor (the operation of the relay is checked; there should be zero voltage across the resistor).

Important - if you connect welding without a current-limiting resistor, an explosion is possible!

1. The use of this type of resistor significantly reduces current surges when welding is connected to a 220V network.
2. Our tool produces a current of more than 100A. This parameter depends on the specific circuit used, and it can be calculated using an oscilloscope.
3. Checking the welding mode on the control unit of a homemade plasma cutter. To do this, you need to connect a voltmeter to the output of the optocoupler amplifier. For low-power devices, the average amplitude voltage should be about 15V.
4. Next, you need to check the output bridge for correct assembly. To do this, a 16V voltage is supplied from a suitable power supply to the input of the unit. The unit at idle speed consumes a current of about 100 mA, which is worth taking into account when performing control measurements.
5. The operation of your homemade inverter can be compared with the operation of an industrial one. On both windings, an oscilloscope measures the correspondence of the pulses to each other.
6. Next you need to check the work. It is necessary to change the voltage from 16V to 220V, connecting the inverter directly to the mains. Using an oscilloscope connected to the output transistors, we observe the signal shape and its compliance with tests at the minimum voltage.

An inverter for welding is a fairly popular unit in any field of activity: in production, at home. And thanks to the use of a built-in regulator and current rectifier, an inverter-type welding unit will allow you to achieve the most effective welding results when compared with the results of similar work using standard welding units on which electrical steel transformers are installed.

Conclusion

Assembling a homemade one is not particularly difficult. If you do not have sufficient experience for this, you can always turn to specialists for additional advice. But as a result, you can assemble a unit with additional functions, which factory analogues lack, and significantly save money.

Inverter welding machines are widely used in the construction industry due to their high performance and low weight. However, not everyone can afford such a tool. The only way out is to make a welding inverter with your own hands. There are many diagrams of such devices on the Internet. Many of them are complex and high-cost, but there are also budget models.

General information about the welding inverter

Traditional welding machines have enough low price, easy repairability, but a very significant drawback is not only their weight, but also their dependence on voltage. The input of the electronic meter is limited to a power of 4 to 5 kW. To weld thick metal, the machine consumes significant power and often the work becomes impossible. They were replaced by inverter welding machines.

Purpose and operating features

Used for carrying out welding work at home, as well as in enterprises, about ensures stable combustion and maintaining the welding arc using high frequency current (other than 50 Hz).

The welding inverter is an ordinary switching power supply, the operation of which is based on the following principles:

1. The input voltage (mains power supply to the inverter welding machine is 220 V AC) is converted to DC.
2. Direct current is converted into high-frequency alternating current.
3. The process of voltage conversion occurs by reducing it.
4. Current rectification and conversion for welding operations while maintaining frequency.

Thanks to these points, the weight and dimensions of the device are reduced. In order to assemble inverter welding with your own hands, you need to know the operating principle of this device.

Operating principle of the equipment

In previous models, the main element was a huge powerful power transformer, which made it possible to obtain powerful currents in the secondary winding, necessary for welding work. To obtain such a current, it is necessary to use a wire with a large diameter, which affects the weight of the welding machine.

With the invention of a switching power supply, it turned out to be easier to solve the problem with weight and size, because the size and weight of the transformer itself are reduced by several tens or hundreds of times. For example, by increasing the frequency by 6 times, you can reduce dimensions transformer and 3 times. This leads to significant material savings.

Thanks to the powerful key transistors used in the inverter circuit, switching occurs with a frequency of 50 to 80 kHz. These transistors operate only on constant voltage.

As you know from a physics course, to obtain a constant voltage, the simplest semiconductor device is used - a diode. The diode passes current in one direction, cutting off negative values ​​of the sinusoidal voltage. But the use of one diode leads to large losses, so a group consisting of powerful diodes is used, which is called a diode bridge.

The output of the diode bridge produces a constant pulsating voltage. To obtain normal DC voltage, a capacitor filter is used. After these transformations, a DC voltage of over 220 V appears at the filter output.

A block consisting of a rectifier bridge and filter elements is called a power supply unit (PSU).

The power supply serves as the power source for the inverter circuit. The transistors are connected to a step-down transformer, which is pulsed and operates at frequencies in the range from 50 to 90 kHz. The power of such a transformer is approximately the same as that of its huge brother - the welding power transformer.

Modernization of such a device becomes lighter, because due to its size and weight, there are additional opportunities to increase the stability of the welding machine.

There is a huge number of home-made welding inverters, the circuits of which vary in functionality and installation methods. Let's analyze each of the homemade models in detail.

Manufacturing a resonant inverter

As a basis, you need to use an AT form factor computer power supply, which will require a cooler and radiators. Parts are taken from the basic base of monitors and televisions, otherwise, if they are not available, they are bought on the market. All components are low cost.

Then you need to decide on the parameters of inverter welding with your own hands. It is also possible to use the following characteristics:

Equipment diagram

The main part - the master oscillator is assembled on the SG3524 chip, which is used in all sources uninterruptible power supply. The inverter has a low power consumption of about 2.5 kW, making it possible to use in an apartment.

The transformer must be assembled and cores of type E42, which is used in old lamp monitors. For production you need approximately 5 pieces of such transformers.

Another transformer should be used for the choke. The remaining inductance elements are assembled from a 2000NM type core. Diodes and transistors must be installed on radiators with KTP-8 or another type of thermal paste. The open circuit voltage is approximately 36 V with a long arc of 4 to 5 mm, which allows novice builders to work with it. The output cables should be laid in ferrite tubes or ferrite rings of the power supply.

A design feature of the circuit is the occurrence of a maximum current in winding I during resonance.

Scheme 1 - Scheme of a welding resonant inverter

Thanks to its low weight and dimensions, it becomes possible to modernize the device.

Preventing Electrode Sticking

For this case, the IRF510 transistor, which is a field effect transistor, is used. In addition, it provides more smooth start and input interrupt on the SG3524 chip:

1. When the temperature is high, the temperature sensor is triggered.
2. Switch off using a toggle switch.
3. Blocking in case of short circuit (short circuit).

Simple welding device

This model is designed for a voltage of 220 V and a current of 32A; after conversion, its value will reach 280A. This value is quite enough for a strong seam at a distance of up to 1.5 centimeters.

Diagram and components

The main element is a transformer, which is quite difficult to make, but quite doable.

Basic data:

1. Consists of a ferrite core (7x7 or 8x8).
2. The primary winding is approximately 100 turns and its diameter is 0.3 mm.
3. Secondary windings - 3 pieces: 15 turns and wire diameter 1 mm; 15 turns - 0.2 mm; 20 turns - 0.35 mm.
4. Materials for the transformer: copper wires of the appropriate diameter, fiberglass, textolite, electrical steel (for iron ore), cotton material.

To clearly understand the principle of operation, it is necessary to carefully study the diagram of the main components.

Figure 1 - Block diagram of an inverter welding machine

Explanation of the diagram:

Power supply and power section

The block, consisting of a transformer, rectifier and filter (or filter system), is made separately from the power section.

Scheme 2 - Schematic diagram of the power supply

Conductors (no more than 15 cm long) to control the gates of transistors must be soldered closer to the latter, and the conductors are connected in pairs to each other, their cross-section does not matter.

The basis of the power unit is a step-down transformer with a Ш20×208 2000 nm core, and winding II is wound in several layers of wire, the insulation of which is not damaged. The secondary must be wound in the following way, isolating the layers: 3 layers, and then a fluoroplastic gasket, then again 3 layers and again a fluoroplastic gasket. This is done to increase overload resistance. Then place a capacitor of at least 1000 V on winding II.

To ensure air circulation between the layers of windings, it is necessary to assemble a current transformer connected to the positive on a ferrite core, and its core should be wrapped with thermal paper (cash tape). Attach the rectifier diodes to the radiator.

Diagram 3 - Power part of the inverter

Inverter unit and cooling

The main purpose of the inverter unit is the process of converting direct to alternating high-frequency current. Are used for this powerful transistors, although in some cases it is possible to replace a more powerful one with 2 or more medium-power transistors.

An important element of the entire device is fairly good cooling. To do this, use a cooler with computer equipment, but should not be limited to one, because it is necessary to provide sufficient cooling for the power circuit, the radiators of which serve to remove heat, but this heat must be dissipated. For complete protection, it is necessary to install a temperature sensor (installed on heating element), thanks to which the power from the network will be disconnected.

Soldering, configuration and performance testing

The key factor is soldering, because when correct placement parts depends on the size of the entire product and the possibility of optimal cooling. Diodes and transistors are installed in opposite directions to each other. The input circuit is designed with a margin of approximately 300 V.

To configure the operation you need connect the pulse width modulator to 15 V to power the cooler. The relay is turned on together with resistor R11 and should produce 150mA.

After these manipulations, you need to proceed directly to checking the functionality of the device:

If this circuit seemed very complicated, then consider the circuit of a very simple device.

The simplest inverter device for welding

The model of this unit is very simple and budget-friendly. It is easy to assemble thanks to a simple circuit diagram.

The entire assembly process can be divided into stages; in addition, it is necessary to collect all the parts and materials:

Scheme 4 - Scheme of the simplest DIY welding inverter

After assembly, the device must be configured and diagnostics performed upon first startup to identify operational errors.

Inverter setting:

Thus, you can assemble an inverter for welding with your own hands. Optional complex circuits, because radio amateurs have found the optimal solution in a budget option. And the level of complexity of the schemes varies from quite complex to simple. To assemble a welding inverter with your own hands, it is not necessary to buy expensive parts, but you can use improvised means.

Welding machines have become part of the everyday life of home craftsmen. Traditional transformers are inexpensive, easy to repair, and this design can be made by hand.

However, they have a drawback - to weld metal thicker than a car body, high currents are required. This gives a load on the side of the primary winding of 220 volts, about 3-5 W.

It will not be possible to weld a pipe in an apartment, because technical specifications, the meter input is limited to 3.5-5 W. And in a private house, power loss is guaranteed.

For work in domestic conditions, it is better to use a welding inverter. This device has less power, compact dimensions and light weight.

The cost of such a machine is higher than that of a conventional transformer machine. Therefore, many homemade “kulibins” are made with their own hands.

Unlike a transformer, in the manufacture of which you struggle with the large weight and thickness of the secondary winding, the inverter offers a solution to other problems.

The circuit of a welding inverter can shock even an experienced radio amateur, not to mention a home handyman whose knowledge is limited to replacing a fuse.


Don't be afraid. Following the assembly instructions, any radio amateur who knows how to hold a soldering iron in his hands will assemble this unit in a few free evenings.

Important! During operation, the welding inverter uses high-frequency currents, so some elements become very hot.

1. A little theory and basic requirements for a welding machine.

Due to the fact that this manual is not technological map, then I do not present the layout of printed circuit boards, nor the design of radiators, nor the order of placement of parts in the case, nor the design of the case itself! All this does not matter and does not affect the operation of the device in any way! The only important thing is that about 50 watts are allocated on the transistors (all together, not just one) of the bridge, and about 100 watts on the power diodes, too, for a total of about 150 watts! I don’t care much about how you use this heat, even if you put them in a glass of distilled water (just kidding :-))), the main thing is not to heat them above 120 degrees C. Well, we’ve sorted out the design, now a little theory and you can start setting it up.
What is a welding machine - it is a powerful power supply capable of operating in the mode of formation and continuous burning of an arc discharge at the output! This is a rather heavy duty mode and not every power supply can work in it! When the end of the electrode touches the metal being welded, a short circuit occurs in the welding circuit; this is the most critical mode of operation of the power supply unit (PSU), since heating, melting and evaporation of a cold electrode requires much more energy than simple arc burning, i.e. The power supply must have a power reserve sufficient for stable ignition of the arc, when using an electrode of the maximum diameter allowed for this device! In our case it is 4mm. An ANO-21 type electrode with a diameter of 3 mm burns stably at currents of 110-130 amperes, but if this is the maximum current for a power supply unit, then lighting the arc will be very problematic! For stable and easy arc ignition, another 50-60 amperes are needed, which in our case is 180-190 amperes! And although the ignition mode is short-term, the power supply must withstand it. Let's go further, the arc has ignited, but according to the laws of physics, the current-voltage characteristic (CVC) of an electric arc in air, at atmospheric pressure, when welding with a coated electrode, has a falling appearance, i.e. The greater the current in the arc, the lower the voltage on it, and only at currents greater than 80A does the arc voltage stabilize and remain constant as the current increases! Based on this, it can be understood that for easy ignition and stable combustion of the arc, the current-voltage characteristic of the power supply must intersect twice with the arc-voltage characteristic! Otherwise, the arc will not be stable with all the ensuing consequences, such as lack of penetration, porous seam, burns! Now we can briefly formulate the requirements for power supply;
a) taking into account the efficiency (about 80-85%), the power of the power supply must be at least 5 kW;
b) must have smooth adjustment of the output current;
c) at low currents it is easy to ignite an arc, have a hot ignition system;
d) have overload protection when the electrode sticks;
e) output voltage at xx is not lower than 45V;
e) complete galvanic isolation from the 220V network;
g) falling current-voltage characteristic.
That's all! The device I developed meets all these requirements, the technical characteristics and electrical diagram of which are given below.

2. Technical characteristics of a homemade welding machine

Supply voltage 220 + 5% V
Welding current 30 - 160 A
Rated arc power 3.5 kVA
Open circuit voltage at 15 turns in the primary winding 62 V
Duty cycle (5 min.),% At max current 30%
PV at a current of 100A 100% (the given PV applies only to my device, and completely depends on cooling, the more powerful the fan, the greater the PV) Maximum consumed
current from the network (measured by constant) 18 A
Efficiency 90%
Weight including cables 5 kg
Electrode diameter 0.8 - 4 mm

The welding machine is designed for manual arc welding and shielding gas welding on direct current. High quality of welds is ensured by additional functions performed in automatic mode: with RDS
- Hot start: from the moment the arc is ignited, the welding current is maximum for 0.3 seconds
- Stabilization of arc combustion: at the moment the drop detaches from the electrode, the welding current automatically increases;
- In case of a short circuit and sticking of the electrode, overload protection is automatically activated; after the electrode is torn off, all parameters are restored after 1s.
- When the inverter overheats, the welding current gradually decreases to 30A, and remains so until it cools completely, then automatically returns to the set value.
Complete galvanic isolation provides 100% protection of the welder from electric shock.

3. Schematic diagram of a resonant welding inverter

Power block, swing block, protection block.
Dr.1 - resonant choke, 12 turns on 2xW16x20, PETV-2 wire, diameter 2.24, gap 0.6mm, L=88mkH Dr.2 - output choke, 6.5 turns on 2xW16x20, PEV2 wire, 4x2.24 , gap Zmm, L=10mkH Tr. 1 - power transformer, primary winding 14-15 turns PETV-2, diameter 2.24, secondary 4x(3+3) with the same wire, 2xW20X28, 2000NM, L=3.5mH Tr.2 - current transformer, 40 turns per ferite ring K20x12x6.2000NM, wire MGTF - 0.3. Tr.Z - master transformer, 6x35 turns on a K28x16x9.2000NM ferite ring, MGTF wire - 0.3. Tr.4 - step-down transformer 220-15-1. T1-T4 on the radiator, power diodes on the radiator, 35A input bridge on the radiator. * All timing capacitors are film capacitors with minimal TKE! 0.25x3.2 kV are collected from Yushtuk 0.1x1.6 kV type K73-16V in series-parallel. When connecting Tr.Z, pay attention to the phases; transistors T1-T4 operate diagonally! Output diodes 150EBU04, RC circuits parallel to the diodes are required! With such winding data, the diodes work with overload, it is better to install them two in parallel, the central one is brand 70CRU04.

4. Selection of power transistors

Power transistors are the heart of any welding machine! The reliability of the entire device depends on the correct choice of power transistors. Technological progress does not stand still; many new semiconductor devices appear on the market, and it is quite difficult to understand this variety. Therefore, in this chapter I will try to briefly outline the basic principles for choosing power switches when building a powerful resonant inverter. The first thing you need to start with is an approximate determination of the power of the future converter. I will not give abstract calculations, and will immediately move on to our welding inverter. If we want to get 160 amperes in an arc at a voltage of 24 volts, then by multiplying these values ​​we will get the useful power that our inverter must deliver without burning out. 24 volts is the average burning voltage of an electric arc 6 - 7 mm long; in fact, the length of the arc changes all the time, and accordingly the voltage on it changes, and the current also changes. But for our calculation this is not very important! So, multiplying these values, we get 3840 W, approximately estimating the converter efficiency of 85%, you can get the power that the transistors must pump through themselves, this is approximately 4517 W. Knowing the total power, you can calculate the current that these transistors will have to switch. If we are making a device to operate from a 220 volt network, then simply dividing the total power by the network voltage, we can obtain the current that the device will consume from the network. That's approximately 20 amps! I get a lot of emails asking if it is possible to make a welding machine so that it can run on a 12 volt car battery? I think these simple calculations will help all those who like to ask them. I foresee the question why I divided the total power into 220 volts, and not into 310, which is obtained after rectification and filtering mains voltage, everything is very simple, in order to maintain 310 volts at a current of 20 amperes, we need a filter capacity of 20,000 microfarads! And we set no more than 1000 uF. We seem to have sorted out the current value, but this should not be the maximum current of the transistors we have chosen! Now in the reference data of many companies two maximum current parameters are given, the first at 20 degrees Celsius, and the second at 100! So, with large currents flowing through the transistor, heat is generated on it, but the rate at which it is removed by the radiator is not high enough and the crystal can heat up to a critical temperature, and the more it heats up, the less its maximum permissible current will be, and ultimately this may lead to destruction of the power key. Typically, such destruction looks like a small explosion, in contrast to a voltage breakdown, when the transistor simply burns out quietly. From here we conclude that for an operating current of 20 amperes it is necessary to choose transistors whose operating current will be at least 20 amperes at 100 degrees Celsius! This immediately narrows our search area to several dozen power transistors.
Naturally, having decided on the current, we must not forget about the operating voltage; in a bridge circuit with transistors, the voltage does not exceed the supply voltage, or, more simply put, cannot be more than 310 volts, when powered from a 220 volt network. Based on this, we select transistors with a permissible voltage of at least 400 volts. Many may say that we will set it to 1200 right away, this will supposedly be more reliable, but this is not entirely true, transistors are of the same type, but for different voltages they can be very different! Let me give an example: IGBT transistors from IR type IRG4PC50UD - 600V - 55A, and the same transistors for 1200 volts IRG4PH50UD - 1200V - 45A, and that’s not all the differences, with equal currents on these transistors there is a different voltage drop, on the first 1.65V, and on the second 2.75V! And with currents of 20 amperes, this is an extra watt of loss, moreover, this is power that is released in the form of heat, it must be removed, which means you need to almost double the radiator! And this is not only additional weight, but also volume! And all this must be remembered when choosing power transistors, but this is just the first estimate! The next stage is the selection of transistors according operating frequency, in our case, the transistor parameters must be maintained at least up to a frequency of 100 kHz! There is one little secret: not all companies provide cutoff frequency parameters for operation in resonant mode, usually only for power switching, and these frequencies are at least 4 to 5 times lower than the cutoff frequency when using the same transistor in resonant mode. This slightly expands the area of ​​our search, but even with such parameters there are several dozen transistors from different companies. The most affordable of them, both in price and availability, are transistors from IR. These are mainly IGBTs, but there are also good field-effect transistors with a permissible voltage of 500 volts, they work well in such circuits, but are not very convenient to fasten, there is no hole in the case. I will not consider the parameters for turning on and off these transistors, although these are also very important parameters, I will briefly say that for normal operation of IGBT transistors, a pause between closing and opening is necessary for all processes inside the transistor to be completed, at least 1.2 microseconds! For MOSFET transistors, this time cannot be less than 0.5 microseconds! These are actually all the requirements for transistors, and if all of them are met, then you will get a reliable welding machine! Based on everything stated above - the best choice these are transistors from IR type IRG4PC50UD, IRG4PH50UD, field-effect transistors IRFPS37N50A, IRFPS40N50, IRFPS43N50K. These transistors have been tested and shown to be reliable and durable when operating in a resonant welding inverter. For low-power converters whose power does not exceed 2.5 kW, you can safely use IRFP460.

POPULAR TRANSISTORS FOR PULSE POWER SUPPLY
NAME	VOLTAGE		RESISTANCE	POWER	CAPACITY SHUTTER	Qg (MANUFACTURER)
NETWORK (220 V)
						17...23nC ( ST)
						38...50nC ( ST)
						35...40nC ( ST)
						39...50nC ( ST)
						46nC ( ST)
						50...70nC ( ST)
						75nC ( ST)
						84nC ( ST)
						65nC ( ST)
						46nC ( ST)
						50...70nC ( ST)
						75nC ( ST)
						65nC ( ST)
STP20NM60FP						54nC ( ST)
						150nC(IR) 75nC ( ST)
						150...200nC (IN)
						252...320nC (IN)
						87...117nC ( ST)

5. Description of operation and method of setting up welding machine components.

Let's move on to the electrical diagram. The master oscillator is assembled on the UC3825 chip, this is one of the best push-pull drivers, it has everything, protection for current, voltage, input, output. During normal operation it is practically impossible to burn it! As can be seen from the circuit diagram, this is a classic push-pull converter, the transformer of which controls the output stage.

The master generator of the welding machine is configured as follows: we supply power and drive the frequency-setting resistor into the range of 20-85 kHz, load the output winding of the transformer Tr3 with a 56 Ohm resistor and look at the signal shape, it should be the same as in Fig. 1

Fig.1

The dead time or step for IGBT transistors must be at least 1.2 μs; if MOSFET transistors are used, then the step can be less, approximately 0.5 μs. The step itself is formed by the frequency-setting capacitance of the driver, and with the details indicated in the diagram, this is about 2 μs. This is where we complete the SG setup for now.
The output stage of the power supply unit is a full resonant bridge assembled on IGBT transistors such as IRG4PC50UD; these transistors can operate up to 200 kHz in resonant mode. In our case, the output current is controlled by changing the frequency of the main generator from 35 kHz (maximum current) to 60 kHz (minimum current), and although a resonant bridge is more difficult to manufacture and requires more careful adjustment, all these difficulties are more than compensated by reliable operation and high efficiency, the absence of dynamic losses on the transistors, the transistors switch at zero current, which allows the use of minimal radiators for cooling; another remarkable property of the resonant circuit is self-limiting power. This effect is explained simply, the more we load the output transformer, and it is an active element of the resonant circuit, the more the resonance frequency of this circuit changes, and if the process of increasing the load occurs at a constant frequency, the effect of automatically limiting the current flowing through the load and, naturally, through the entire bridge!
This is why it is so important to tune the device under load, that is, in order to get maximum power in an arc with parameters of 150A and 22-24V, you need to connect an equivalent load to the output of the device, this is 0.14 - 0.16 Ohm, and by selecting the frequency, adjust the resonance, namely at this load the device will have maximum power and maximum efficiency, and then even in short circuit mode (short circuit), despite the fact that a current exceeding the resonant one will flow in the external circuit, the voltage will drop almost to zero, and accordingly the power will decrease and the transistors will will not enter overload mode! And yet, the resonant circuit operates in a sinusoid and the current also increases according to the sinusoidal law, that is, dl/dt does not exceed the permissible modes for transistors, and snubbers (RC chains) are not required to protect transistors from dynamic overloads, or, more understandably, from too steep there will simply be no fronts at all! As we see, everything seems to be beautiful and it seems that the overcurrent protection circuit is not needed at all, or is needed only during the setup process, do not be fooled, because the current is adjusted by changing the frequency, and there is small plot on the frequency response, when resonance occurs during a short circuit, in this place the current through the transistors may exceed the permissible current for them, and the transistors will naturally burn out. And although it is quite difficult to specifically get into this particular mode, according to the law of meanness it is quite possible! This is when you will need current protection!
The volt-ampere characteristic of the resonant bridge immediately has a falling appearance, and naturally there is no need to artificially shape it! Although, if necessary, the angle of inclination of the current-voltage characteristic can be easily adjusted using a resonant choke. And one more property, which I cannot help but talk about, and having learned about it you will forever forget the power switching circuits that are available in abundance on the Internet, this wonderful property is the ability to operate several resonant circuits on one load with maximum efficiency! In practice, this makes it possible to create welding (or any other) inverters of unlimited power! You can create block designs, where each block will be able to operate independently, this will increase the reliability of the entire structure and make it possible to easily replace blocks when they fail, or you can run several power blocks with one driver and they will all work in phase. So the welding machine, built by me according to this principle, easily produces an arc of 300 amperes, with a weight without a body of 5 kg! And this is only a double set; you can increase the power limitlessly!
This was a slight deviation from the main topic, but I hope it provided an opportunity to understand and appreciate all the delights of the full resonant bridge circuit. Now let's get back to setting up!
It is configured as follows: we connect the SG to the bridge, taking into account the phases (transistors operate diagonally), supply 12-25V power, turn on a 100W 12-24V light bulb in the secondary winding of the power transformer Tr1, changing the frequency of the SG we achieve the brightest glow of the light bulb, in our case it is 30 -35kHz is the resonance frequency, then I will try to talk in detail about how a full resonant bridge works.
Transistors in a resonant bridge (as in a linear bridge) operate diagonally, it looks like this: the upper left T4 and the lower right T2 are simultaneously open, at this time the upper right T3 and the lower left T1 are closed. Or vice versa! The operation of a resonant bridge can be divided into four phases. Let's consider what and how happens if the switching frequency of the transistors coincides with the resonant frequency of the circuit Dr.1-Cut.-Tr.1. Suppose transistors T3, T1 are opened in the first phase, the time they remain in the open state is set by the 3G driver, and at a resonant frequency of 33 kHz, it is 14 μs. At this time, current flows through the Cut. - Dr.1 - Tr.1. The current in this circuit first increases from zero to the maximum value, and then, as the capacitor charges, Cut. , decreases to zero. The resonant inductor Dr.1, connected in series with the capacitor, forms sinusoidal fronts. If you connect a resistor in series with the resonance circuit and connect an oscilloscope to it, you can see a current shape resembling a half-cycle of a sine wave. In the second phase, lasting 2 μs, the gates of transistors T1, T3 are connected to ground through a 56 Ohm resistor and the winding of the pulse transformer Tr.3, this is the so-called “dead time”. During this time, the gate capacitances of transistors T1, T3 are completely discharged, and the transistors close. As can be seen from the above, the moment of transition from an open state to a closed state for transistors coincides with zero current, because the capacitor is Cut. already charged and current no longer flows through it. The third phase begins - transistors T2, T4 open. The time they remain in the open state is 14 μs, during which time the Slice capacitor is completely recharged, forming the second half-cycle of the sinusoid. The voltage to which the Cut is recharged depends on the load resistance in the secondary winding of Tr.1, and the lower the load resistance, the greater the voltage on the Cut. With a load of 0.15 Ohm, the voltage across the resonant capacitor can reach 3 kV. The fourth phase begins, like the second, at the moment when the collector current of transistors T2, T4 decreases to zero. This phase also lasts 2 µs. The transistors turn off. Then everything repeats itself. The second and fourth phases of operation are necessary so that the transistors in the bridge arms have time to close before the next pair opens; if the time of the second and fourth phases is less than the time required for the complete closing of the selected transistors, a through current pulse will arise, almost High voltage short circuit, and the consequences are easily predictable; usually the entire arm (upper and lower transistors) burns out, plus the power bridge, plus the neighbor's traffic jams! :-))). For the transistors used in my circuit, the “dead time” should be at least 1.2 μs, but taking into account the spread of parameters, I deliberately increased it to 2 μs.
One more very important thing to remember is that all elements of the resonant bridge affect the resonance frequency and when replacing any of them, be it a capacitor, inductor, transformer or transistors, to obtain maximum efficiency, you need to re-adjust the resonant frequency! In the diagram I have given the inductance values, but this does not mean that by installing a choke or transformer of a different design that has such inductance, you will receive the promised parameters. It's better to do as I recommend. Will be cheaper!
How a resonant bridge works, in general terms, seems to have become clear, now let’s figure out what, and quite important function is performed by the resonant inductor Dr.1
If upon first adjustment the resonance turns out to be much lower than 30 kHz, do not be alarmed! It’s just the ferrite core Dr1., a little different, this can be easily corrected by increasing the non-magnetic gap; below we describe in detail the tuning process and the design nuances of the resonant choke Dr1.
The most important element of the resonant circuit is resonant choke Dr.1, the power delivered by the inverter to the load and the resonance frequency of the entire converter depend on the quality of its manufacture! During the pre-tuning process, secure the throttle so that it can be removed and disassembled to increase or decrease the clearance. The thing is that the ferrite cores I use are always different, and each time I have to adjust the inductor by changing the thickness of the non-magnetic gap! In my practice, in order to obtain identical output parameters, I had to change the gaps from 0.2 to 0.8 mm! It’s better to start with 0.1 mm, find the resonance and at the same time measure the output power; if the resonance frequency is below 20 kHz, and the output current does not exceed 50-70A, then you can safely increase the gap by 2-2.5 times! All adjustments in the throttle must be made only by changing the thickness of the non-magnetic gap! Do not change the number of turns! Use only paper or cardboard as gaskets, never use synthetic films, they behave unpredictably and can melt or even burn! With the parameters indicated in the diagram, the inductance of the inductor should be approximately 88-90 μH, this is with a gap of 0.6 mm, 12 turns of PETV2 wire with a diameter of 2.24 mm. I repeat once again, you can adjust the parameters only by changing the thickness of the gap! The optimal resonance frequency for ferrites with a permeability of 2000NM lies in the range of 30-35 kHz, but this does not mean that they will not work lower or higher, just that the losses will be slightly different. The throttle core must not be tightened with a metal bracket; in the area of ​​the gap, the metal of the bracket will become very hot!
Next is the resonant capacitor, an equally important detail! In the first designs I installed K73 -16V, but you need at least 10 of them, and the design turns out to be quite cumbersome, although quite reliable. Imported capacitors from WIMA have now appeared MKP10, 0.22x1000V- these are special capacitors for high currents, they work very reliably, I install only 4 of them, they practically don’t take up space and don’t get hot at all! You can use capacitors like K78-2 0.15x1000V, you will need 6 of them. They are connected into two blocks of three in parallel, resulting in 0.225x2000V. They work fine and hardly get hot. Or use capacitors designed for use in induction cookers, type MKP from China.
Well, we seem to have figured it out, we can move on to further configuration.
We change the lamp to a more powerful one and a voltage of 110V, and repeat everything from the beginning, gradually raising the voltage to 220 volts. If everything works, turn off the lamp, connect the power diodes and inductor Dr.2. We connect a rheostat with a resistance of 1 Ohm x 1 kW to the output of the device and repeat everything by first measuring the voltage across the load and adjusting the frequency to resonance, at this moment there will be a maximum voltage on the rheostat, and when the frequency changes in any direction, the voltage decreases! If everything is assembled correctly, the maximum voltage across the load will be about 40V. Accordingly, the load current is about 40A. It is not difficult to calculate the power of 40x40, we get 1600 W, then by reducing the load resistance, we adjust the resonance with a frequency-setting resistor, the maximum current can be obtained only at the resonant frequency, for this we connect a voltmeter in parallel with the load and by changing the frequency of the generator we find the maximum voltage. The calculation of resonant circuits is described in detail in (6). At this moment, you can look at the voltage waveform on the resonant capacitor; there should be a correct sinusoid with an amplitude of up to 1000 volts. When the load resistance decreases (power increases), the amplitude increases to 3 kV, but the voltage shape must remain sinusoidal! This is important, if a triangle occurs, it means that the capacitance is broken or the winding of the resonant choke is shorted, both of which are not desirable! At the values ​​indicated in the diagram, the resonance will be about 30-35 kHz (highly dependent on the permeability of the ferrite).
Another important detail, to obtain the maximum current in the arc, you need to adjust the resonance at the maximum load, in our case, to obtain a current in the arc of 150A, the load during adjustment should be 0.14 ohms! (It is important!). The voltage at the load, when setting the maximum current, should be 22 -24V, this is the normal arcing voltage! Accordingly, the power in the arc will be 150 x 24 = 3600 W, this is enough for normal combustion of an electrode with a diameter of 3-3.6 mm. You can weld almost any piece of iron, I welded rails!
The output current is adjusted by changing the frequency of the generator.
As the frequency increases, the following happens, firstly: the ratio of the pulse duration to the pause (step) changes; secondly: the converter goes out of resonance; and the choke from a resonant one turns into a leakage choke, that is, its resistance directly becomes dependent on the frequency, the higher the frequency, the greater the inductive reactance of the choke. Naturally, all this leads to a decrease in the current through the output transformer; in our case, a change in frequency from 30 kHz to 57 kHz causes a change in the current in the arc from 160 A to 25 A, i.e. 6 times! If the frequency is changed automatically, then you can control the arc current during the welding process, the “hot start” mode is implemented on this principle, its essence is that at any value of the welding current, the current will be maximum for the first 0.3 s! This makes it possible to easily ignite and maintain an arc at low currents. The thermal protection mode is also organized to automatically increase the frequency when a critical temperature is reached, which naturally causes a smooth decrease in the welding current to a minimum value without abrupt shutdown! This is important because a crater does not form as if the arc were abruptly interrupted!
But in general, you can do without these bells and whistles, everything works quite stable, and if you work without fanaticism, the device does not heat up more than 45 degrees C, and the arc ignites easily in any mode.
Next, we will consider the overcurrent protection circuit, as mentioned above, it is needed only at the time of setup and at the moment the short circuit mode coincides with resonance, if the electrode gets stuck in this mode! As you can see, it is assembled on a 561LA7, the circuit is a kind of delay line, the turn-on delay is 4 ms, the turn-off delay is 20 ms, the turn-on delay is necessary to ignite the arc in any mode, even when the short-circuit mode coincides with resonance!
The protection circuit is configured for a maximum current in the primary circuit of about 30A; during setup, it is better to reduce the protection current to 10-15A; to do this, replace the 6k resistor with a 15k resistor in the protection circuit. If everything works, try to strike an arc on some paperclip.
Below I will try to explain why the above protection circuit is not effective during normal operation, the fact is that the maximum current flowing in the primary winding of a power transformer completely depends only on the design of the resonant inductor, more precisely on the gap in the magnetic core of this inductor, and so that we did not do this in the secondary winding, the current in the primary cannot exceed the maximum current of the resonant circuit! Hence the conclusion - protection configured for maximum current in the primary winding of the power transformer can only operate at the moment of resonance, but why do we need it at this moment? Just so as not to overload the transistors at the moment when the short-circuit mode coincides with resonance, and naturally in the event that we assume that the resonant circuit and the power transformer burn out at the same time, then of course such protection is necessary, in fact, for this purpose I included it in the circuit from the very beginning started when I was experimenting with different transistors and various designs chokes, transformers, capacitors. And knowing the inquisitive mind of our people, who will not believe what is written and will wind their tr - ry, chokes, install capacitors in a row, I left it, I think it was not in vain! :-))) There is one more important nuance, no matter how you configure the protection, there is only one condition, the 9th leg of the Uc3825 microcircuit should not receive a smoothly increasing voltage, only a fast edge from 0 to +3(5) V, understanding this cost me several power transistors! And one more tip:
- it’s better to start tuning if there is no gap in the resonant choke, this will immediately limit the short-circuit current in the output winding to 40 - 60A, and then gradually increase the gap and, accordingly, the output current! Remembering to adjust the resonance each time, as the gap increases, it will move towards an increase in frequency!
Below are diagrams of temperature protection Fig. 2, hot start and arc combustion stabilizer Fig. 3, although in the latest developments I do not install them and, as thermal protection, I glue 80°-100°C thermal switches onto the diodes and into the winding of the power transformer and connect them everything is in series, and I turn it off with an additional relay high voltage, simple and reliable! And the arc, at 62V at XX, ignites quite easily and softly, but turning on the “hot start” circuit allows you to avoid the short-circuit mode - resonance! It was mentioned above.

Fig.2

Fig.3

Change in the slope of the current-voltage characteristic as a function of frequency, experimentally obtained curves with a gap in the resonant choke of 0.5 mm. When the gap changes in one direction or another, the steepness of all curves changes accordingly. As the gap increases, the current-voltage characteristics become flatter and the arc becomes more rigid! As can be seen from the obtained graphs, by increasing the gap, you can obtain a fairly rigid current-voltage characteristic. And although the initial section will look steeply falling, a power supply with such a current-voltage characteristic can already be used with a semi-automatic C02, if the secondary winding is reduced to 2+2 turns.

6. New developments and description of their work.

Here are diagrams of my latest developments and comments on them.

Figure 5 shows a diagram of a welding inverter with a modified circuit of the protection unit; a Hall sensor of the Ss495 type is used as a current sensor; this sensor has a linear dependence of the output voltage on the force magnetic field, and inserted into a sawn permalloy ring, allows you to measure currents up to 100 amperes. A wire is passed through the ring, the circuit of which needs protection, and when the maximum permissible current in this circuit is reached, the circuit will give a command to turn off. In my circuit, when the maximum permissible current in the protected circuit is reached, the master oscillator is blocked. I passed a high voltage positive wire (+310V) through the ring, thereby limiting the current of the entire bridge to 20 - 25A. To ensure that the arc ignites easily and the protection circuit does not give false shutdowns, an RC circuit is introduced after the Hall sensor, by changing the parameters of which you can set a delay for turning off the power unit. That’s actually all the changes, as you can see, I practically didn’t change the power part, it turned out to be very reliable, I only reduced the input capacitance from 1000 to 470 microfarads, but this is already the limit, it’s not worth setting less. And without this capacity, I don’t recommend turning on the device at all, high-voltage surges occur and the input bridge can burn out, with all the ensuing consequences! I recommend installing a 1.5KE250CA transil in parallel with the middle diode, in parallel to the diodes RC chains, increase the resistor power to 5 W. The starting system has been changed, now it is also protection against long-term short circuit mode, when the electrode sticks, a capacitor connected in parallel with the relay sets a shutdown delay. If the output has one 150EBU04 power diode per arm, then I recommend not setting more than 50mF, and although the delay will only be a few tens of milliseconds, this is quite enough to ignite the arc and the diodes will not have time to burn out! When connecting two diodes in parallel, you can increase the capacitance to 470mF, and accordingly the delay will increase to several seconds! The starting system works like this: when connected to an alternating current network, an RC circuit consisting of a capacitor with a capacity of 4mF and a resistor with a resistance of 4-6 Ohms limits the input current to 0.3A, the main capacity is 470gg^x350y, it charges slowly and naturally the output voltage increases , as soon as the output voltage reaches approximately 40V, the triggering relay is triggered, closing the RC circuit with its contacts, after which the output voltage rises to 62V. But any relay has an interesting property: it operates at one current, and releases the armature at another current. Usually this ratio is 5/1, to make it clearer, if the relay turns on at a current of 5mA, it will turn off at a current of 1mA. The resistance connected in series with the relay is selected so that it turns on at 40V and turns off at 10V. Since the relay chain - a resistor - is connected parallel to the arc, and as we know, the arc burns in the range of 18 - 28V, then the relay is in the on state, if a short circuit occurs at the output (electrode sticking), then the voltage drops sharply to 3-5V, taking into account the drop on the cables and electrode. At this voltage, the relay can no longer be kept in the on state and opens the power circuit, the RC circuit is turned on, but as long as the short-circuit mode remains in the output circuit, the power relay will be open. After eliminating the short-circuit mode, the output voltage begins to increase, the power relay is activated and the device is ready for operation again, this whole process takes 1-2 seconds, and is practically unnoticeable, and after tearing off the electrode, you can immediately begin new attempts to ignite the arc. :-))) Usually the arc does not ignite well if the current is selected incorrectly, the electrodes are damp or of poor quality, or the coating is sprinkled. In general, it should be remembered that welding DC, if the voltage does not exceed 65V, ideally dry electrodes are required! Usually on the packaging of electrodes they write the voltage XX for welding at direct current at which the electrode should burn stably! For ANO21, the XX voltage must be more than 50 Volts! But this is for calcined electrodes! And if they were stored for years in a damp basement, then naturally they will burn poorly, and it is better if the XX voltage is higher. With 14 turns in the primary winding, the idle voltage is about 66V. At this voltage, most electrodes burn normally.
To also reduce weight, instead of a 15V transformer, a converter on the IR53HD420 chip was used; this is a very reliable chip, and it is easy to create a power supply with a power of up to 50W. The transformer in the power supply is wound in a B22 - 2000NM cup, the primary winding is 60 turns, PEV-2 wire, 0.3 mm in diameter, the secondary is 7+7 turns, wire with a diameter of 0.7 mm. The conversion frequency is 100 -120 kHz, I recommend installing a trimmer as a frequency-setting resistor, so that in case of beats with the power unit, you can change the frequency! The appearance of beats means the death of the device!

Throttle design Dr.1 and dr.2

Cardboard spacers, 3 pcs. For Dr.1 0.1 - 0.8 mm (selected during setup) for Dr.2 - 3 mm.
Core 2xW16x20 2000NM
The coil frame is glued together from thin fiberglass, placed on a wooden frame, and the required number of turns is wound. Dr.1 - 12 turns, PETV-2 wire, diameter 2.24 mm, wound with an air gap between turns, gap thickness 0.3 - 0.5 mm. You can use a thick cotton thread, carefully laying it between the turns of the wire, see the picture. Dr.2 - 6.5 turns wound into four wires, brand PETV -2, diameter 2.24 mm, total cross-section 16 sq. , is wound closely, in two layers. The coils need to be fastened, using epoxy resin.

Fig. 6 design of the resonant and output choke.

Fig. 7 shows the design of the power unit, a kind of “layer cake”, this is for the lazy :-)))

Fig.8

Fig.9

Fig.10

Fig.11

Fig. 8 - 11 wiring of the control unit, for those who are generally confused about everything :-))). Although it is necessary to figure out what leads where and where!

Hot start scheme

Fig. 12 Soft ignition circuit

Fig. 12 soft ignition system, very effective when operating at low currents. It is practically impossible not to strike an arc, you just place the electrode on the metal and gradually begin to withdraw, a low-ampere arc appears, it cannot weld the electrode, there is not enough power, but it burns and stretches perfectly, lights like a match, very beautiful! Well, when this arc lights up, the power one is connected in parallel; if suddenly the electrode gets stuck, then the power current is instantly switched off, leaving only the ignition current. And until the arc lights up, the power current does not turn on! I advise you to install it, the arc will be under any conditions, the power unit is not overloaded and always works in optimal mode, short-circuit currents are practically eliminated!

Fig.13

The power arc control unit is shown in Fig. 13. It works like this - it measures the voltage at the output resistor of the ignition system, and gives a signal to start the power unit only in the voltage range 55 - 25V, that is, only at the moment when the arc is burning!

Relay P contacts operate to close and are connected to the break in the high-voltage circuit of the power unit. Relay 12VDC, 300VDC x 30A.
It is quite difficult to find a relay with such parameters, but you can go the other way :-)) turn the relay to open, connect one contact to +12V, and the second through a 1kOhm resistor, connect to the 9th leg of the Uc3825 microcircuit in the ZG block. It works just as well! Or apply the diagram below in Fig. 15,

The circuit is completely autonomous, but with simple modifications, it can be used simultaneously as a power supply (12V) for the control circuit, the power of this converter is no more than 200W. It is necessary to install radiators on transistors and diodes. When connecting "MP", output capacitors and output choke in the power unit should be completely excluded. Figure 14 shows a complete diagram of a welding inverter with a soft ignition system.

the connection point is shown with a red dotted line in Fig. 14

Fig. 16. Working diagram of one of the options for soft arson

7. Conclusion

In conclusion, I would like to briefly note the main points that need to be remembered when designing a powerful resonant welding inverter:
a) completely eliminate PWM, for this you need a stabilized supply voltage for the master oscillator, no changing voltages at the inputs of the “error” amplifier (1,3), the minimum “soft start” time is set by the capacitance at (8), blocking the microcircuit (9) only a sharp voltage drop, best logical from 0 to +5V with a steep rising edge, switching on by the same logical decline from +5V to 0;
b) it is imperative to install two-anode zener diodes of type KS213 in the gates of power transistors;
c) place the control transformer in close proximity to the power transistors, twist the wires going to the gates in pairs;
d) when wiring the power bridge board, remember that significant currents will flow along the tracks (up to 25A), so the (-) bus and (+) bus, as well as the busbars for connecting the resonant circuit, must be made as wide as possible, and the copper must be tinned;
e) all power circuits must have reliable connections, it is best to solder them; poor contact, with currents greater than 100A, can lead to melting and fire of the internal parts of the device;
f) the network connection wire must have a sufficient cross-section of 1.5 - 2.5 mm sq;
g) be sure to install a 25A fuse at the input, you can install a machine;
h) all high-voltage circuits must be reliably isolated from the housing and output;
i) do not tighten the resonant choke with a metal bracket or cover it with a solid metal casing;
j) it must be remembered that a significant amount of heat is generated on the power elements of the circuit; this must be taken into account when placing parts in the housing; it is necessary to provide a ventilation system;
k) it is imperative to install protective RC circuits in parallel with the output power diodes; they protect the output diodes from voltage breakdown;
m) never use any garbage as a resonant capacitor, this can lead to very disastrous results, only those types that are indicated in the diagram are K73-16V (0.1x1600V) or WIMA MKP10 (0.22x1000V), K78-2 ( 0.15x1000V) by connecting them in series and in parallel.
Strict adherence to all of the above points will ensure 100% success and your safety. You must always remember - power electronics does not forgive mistakes!

8. Schematic diagrams and a description of the operation of an inverter with a leakage choke.

One of the ways to create a falling volt-ampere characteristic of a welding machine is to use a leakage choke. The Fast and the Furious apparatus was built according to this scheme. This is something between an ordinary bridge, the current in which is controlled by PWM, and a resonant bridge, controlled by a change in frequency.

I will try to highlight all the pros and cons of this construction of a welding inverter. Let's start with the advantages: a) current regulation is frequency-based; as the frequency increases, the current decreases. This makes it possible to regulate the current in automatic mode, making it easy to build a “hot start” system.
b) the falling current-voltage characteristic is formed by a leakage inductor, this construction is more reliable than parametric stabilization with PWM, and faster, there is no delay for turning on the active elements. Simplicity and reliability! Perhaps these are all advantages. :-(^^^L
Now about the disadvantages, there are not many of them either:
a) transistors operate in linear switching mode;
b) snubbers are required to protect transistors;
c) narrow current adjustment range;
d) low conversion frequencies, due to the parameters of power switching of transistors;
but they are quite significant, and require their own methods of compensating for them. Let's analyze the operation of an inverter built on this principle, see Fig. 17 As you can see, its circuit is practically no different from the circuit of a resonant inverter, only the parameters of the LC chain in the diagonal of the bridge have been changed, snubbers have been introduced to protect transistors, the resistance of resistors connected in parallel to the gate windings of the master transformer has been reduced, and the power of this transformer has been increased.
Let's consider an LC circuit connected in series with a power transformer, the capacitance of capacitor C has been increased to 22 μR, now it works as a balancing capacitor that prevents the core from being magnetized. The short-circuit current of the converter, the power adjustment range, and the conversion frequency of the inverter completely depend on the parameters of the inductor L. At the conversion frequencies of the Fast and Furious 125 device, which is 10 - 50 kHz, the inductance of the inductor is 70 μH, at a frequency of 10 kHz the resistance of such an inductor is 4.4 Ohms, therefore the short-circuit current through the primary circuit will be 50 amperes! But not more! :-) For transistors, this is of course a bit much, so Fast and the Furious uses two-stage overcurrent protection, limiting the short-circuit current to 20-25 amperes. The current-voltage characteristic of such a converter is a steeply falling straight line, linearly dependent on the output current.
As the frequency increases, the reactance of the inductor increases, therefore, the current flowing through the primary winding of the output transformer is limited, and the output current decreases linearly. The disadvantage of such a current control system is that the shape of the current with increasing frequency becomes similar to a triangle, and this increases dynamic losses, and excess heat is generated on the transistors, but given that the total power decreases and the current through the transistors also decreases, these values ​​can be neglected.
In practice, the most significant drawback of an inverter circuit with a leakage choke is the operation of transistors in the mode of linear (power) current switching. Such switching places increased demands on the driver that controls these transistors. It is best to use drivers on IR microcircuits, which are directly designed to control the upper and lower switches of the bridge converter. They produce clear pulses into the gates of controlled transistors, and unlike a transformer control system, they do not require much power. But the transformer system forms a galvanic isolation, and if the power transistors fail, the control circuit remains operational! This is an undeniable advantage not only from the economic side of building a welding inverter, but also from the side of simplicity and reliability. Figure 18 shows a circuit diagram of an inverter control unit with drivers, and Figure 17 shows control via a pulse transformer. The output current is regulated by changing the frequency from 10 kHz (Imax) to 50 kHz (1t1p). If you install higher-frequency transistors, the range of current adjustments can be slightly expanded.
When constructing an inverter of this type, it is necessary to take into account exactly the same conditions as when constructing a resonant converter, plus all the features of constructing a converter operating in linear switching mode. This is: strict stabilization of the supply voltage of the master unit, the mode of PWM occurrence is unacceptable! And all other features listed in paragraph 7 on page 31. If drivers on microcircuits are used instead of a control transformer, always remember that the minus of the low-voltage supply will be connected to the network, and take additional safety measures!

Control unit on IR2110

Fig.18

9. Design and circuit solutions proposed and tested
my friends and followers.

1. Power transformer wound on one core type Sh20x28 2500NMS, primary winding 15 turns, PETV-2 wire, diameter - 2.24 mm. Secondary 3+3 turns wire 2.24 in four wires, total cross-section 15.7 mm sq.
It works well, the windings practically do not heat up even at high currents, and easily discharges more than 160A into the arc! But the core itself heats up, up to about 95 degrees, you need to put it in airflow. But on the other hand, weight is gained (0.5 kg) and volume is freed up!
2. The secondary winding of the power transformer is wound with copper tape 38x0.5mm, core 2Ш20x28, primary winding 14 turns, PEV-2 wires, diameter 2.12.
It works great, the voltage is about 66V, it heats up to 60 degrees.
3. The output choke is wound on one Ш20х28, 7 turns of stranded copper wire, with a cross-section from 10 to 20 mm sq., does not affect the work in any way. Gap 1.5 mm, inductance 12 μH.
4. Resonance choke - wound on one Ш20х28, 2000НМ, 11 turns, PETV2 wire, diameter 2.24. The gap is 0.5mm. Resonance frequency 37 kHz.
Works good.
5. Instead of Uc3825, 1156EU2 was used.
Works great.
6. The input capacitance varied from 470 µF to 2000 µF. If the gap does not change
in a resonant choke, then with an increase in the capacitance of the input capacitor, the power supplied to the arc increases proportionally.
7. Current protection was completely eliminated. The device has been working for almost a year and is not going to burn out.
This improvement simplified the scheme to the point of complete shamelessness. But the use of protection against long-term short circuit and the “hot start” + “non-stick” system almost completely eliminates the occurrence of current overload.
8. The output transistors are placed on one radiator through silicone-ceramic gaskets, "NOMAKON" type.
They work great.
9. Instead of 150EBU04, two 85EPF06 were installed in parallel. Works great.
10. The current regulation system has been changed, the converter operates at a resonant frequency, and the output current is adjusted by changing the duration of the control pulses.
I checked it, it works great! The current is adjustable practically from 0 to max! The diagram of the device with such adjustment is shown in Fig. 21.

Tr.1 - power transformer 2Ш20х28, primary - 17 turns, ХХ=56V D1-D2 - HER208 D3,D5 - 150EBU04
D6-D9 - KD2997A
P - starting relay, 24V, 30A - 250VAC
Dr.3 - swings on a ferrite ring K28x16x9, 13-15 turns
installation wire with a cross section of 0.75 mm square. Inductance no less
200µN.

The circuit shown in Fig. 19 doubles the output voltage. Double the voltage is applied parallel to the arc. This inclusion facilitates ignition in all operating modes, increases the stability of the arc (the arc easily stretches up to 2 cm), improves the quality of the weld, you can weld with large-diameter electrodes at low currents without overheating the part being welded. Allows you to easily dose the amount of deposited metal; when the electrode is withdrawn, the arc does not go out, but the current decreases sharply. At increased voltage, electrodes of all brands ignite and burn easily. When welding with thin electrodes (1.0 - 2.5 mm) at low currents, ideal quality of the weld is achieved, even for “dummies”. I was able to use a four-piece to weld a 0.8mm thick sheet to a 5mm thick corner (52x52). The XX voltage without doubling was 56V, with a doubler 110V. The doubler current is limited by capacitors of 0.22x630V type K78-2, at the level of 4 - 5 Amps in arc mode, and up to 10A during short circuit. As you can see, we had to add two more diodes for the triggering relay; with this connection, it also protects against long-term short-circuit mode, as in the circuit in Fig. 5. The output choke Dr.2 turned out to be unnecessary, and this is 0.5 kg! The arc burns steadily! The originality of this circuit lies in the fact that the double voltage phase is rotated 180 degrees relative to the power voltage, so the high voltage after the output capacitors are discharged does not block the power diodes, but fills the gaps between pulses with double voltage. It is this effect that increases the stability of the arc and improves the quality of the seam!
The Italians use similar schemes in industrial portable inverters.

Figure 20 shows a diagram of a welding inverter with the most advanced configuration. Simplicity and reliability, a minimum of parts; below are its technical characteristics.

1. Supply voltage 210 -- 240 V
2. Arc current 20 - 200 A
3. Current consumed from the network 8 - 22 A
4. Voltage XX 110V
5. Weight without housing less than 2.5 kg

As you can see, the circuit in Fig. 20 is not very different from the circuit in Fig. 5. But this is a completely finished circuit; it practically does not require additional ignition and arc stabilization systems. The use of an output voltage doubler made it possible to eliminate the output choke, increase the output current to 200A and significantly improve the quality of welds in all operating modes, from 20A to 200A. The arc ignites very easily and pleasantly, electrodes of almost all types burn steadily. When welding stainless steels, the quality of a weld made with an electrode is not inferior to a weld made in argon!
All winding data are similar to the previous designs, only in a power transformer you can wind the primary winding of 17-18 turns using 2.0-2.12 PETV-2 or PEV-2 wire. Now there is no point in increasing the output voltage of the transformer, 50-55V is enough for excellent operation, the doubler will do the rest. The resonant choke is of exactly the same design as in the previous circuits, only it has an increased non-magnetic gap (selected experimentally, approximately 0.6 - 0.8 mm).

Dear readers, several schemes are offered to your attention, but in fact they are the same power point with various additions and improvements. All circuits have been tested many times and have shown high reliability, unpretentiousness and excellent results when operating in various climatic conditions. To make a welding machine, you can take any of the above diagrams, use the proposed changes and create a machine that fully satisfies your requirements. Without changing practically anything, just increasing or decreasing the gap in the resonant choke, increasing or decreasing radiators on the output diodes and transistors, increasing or decreasing the power of the cooler, you can get a whole series of welding machines with a maximum output current from 100A to 250A and duty cycle = 100 %. PV depends only on the cooling system, and the more powerful the fans used and the larger the area of ​​the radiators, the longer your device can operate in continuous mode at maximum current! But an increase in radiators entails an increase in the size and weight of the entire structure, so before you start making a welding machine, you always need to sit down and think for what purpose you will need it! As practice has shown, there is nothing super complicated in designing a welding inverter using a resonant bridge. It is the use of a resonant circuit for this purpose that makes it possible to 100% avoid problems associated with the installation of power circuits, and when manufacturing a power device at home, these problems always arise! The resonant circuit solves them automatically, preserving and extending the life of power transistors and diodes!

10. Welding machine with phase control of output current

The scheme presented in Fig. 21 is the most attractive from my point of view. Tests have shown the high reliability of such a converter. This circuit takes full advantage of the resonant converter, since the frequency does not change, the power switches are always turned off at zero current, and this is an important point from the point of view of the controllability of the switches. The current is adjusted by changing the duration of the control pulses. This circuit solution allows you to change the output current practically from 0 to the maximum value (200A). The adjustment scale is completely linear! Changing the duration of control pulses is achieved by applying a varying voltage in the range of 3-4V to the 8th leg of the Uc3825 microcircuit. Changing the voltage on this leg from 4V to 3V gives a smooth change in the cycle duration from 50% to 0%! Adjusting the current in this way allows you to avoid such an unpleasant phenomenon as the coincidence of resonance with the short-circuit mode, which is possible with frequency regulation. Therefore, another possible overload mode is eliminated! As a result, you can completely remove the current protection circuit by once adjusting the maximum output current by the gap in the resonant choke. The device is configured exactly like all previous models. The only thing that needs to be done is, before starting tuning, set the maximum cycle duration by setting the voltage on leg 8 to 4V; if this is not done, the resonance will be shifted, and maximum power The switching point of the keys may not coincide with zero current. With large deviations, this can lead to dynamic overload of power transistors, their overheating and failure. The use of a voltage doubler at the output makes it possible to reduce the load on the core by increasing the number of turns of the primary winding to 20. The output voltage XX is 46.5 V, respectively, after the doubler 93 V, which meets all safety standards for inverter welding sources! Lowering the output voltage of the power unit allows the use of lower voltage (cheaper) output diodes. You can safely put 150EBU02 or BYV255V200. Below are the wiring data for my latest model welding inverter.
Tr.1 Wire PEV-2, diameter 1.81 mm, number of turns -20. The secondary winding is 3+3, 16mm kV, wound in 4 wires with a diameter of 2.24. The design is similar to the previous ones. Core E65, No. 87 from EPKOS. Our approximate analogue is 20x28, 2200NMS. One core!
Dr.1 10 turns, PETV-2 with a diameter of 2.24 mm. Core 20x28 2000NM. The gap is 0.6-0.8mm. Inductance 66 µH for max current in arc 180-200A. Dr.3 12 turns of installation wire, cross-section 1 mm kV, ring 28x16x9, without gap, 2000NM1
With these parameters, the resonant frequency is about 35 kHz. As can be seen from the diagram, there is no current protection, no output choke, no output capacitors. The power transformer and resonant choke are wound on single cores of type Ш20х28. All this made it possible to reduce weight and free up volume inside the case, and as a result, ease the temperature regime of the entire device, and calmly increase the current in the arc to 200A!

List of useful literature.

1. "Radio" No. 9, 1990
2. "Microcircuits for switching power supplies and their application", 2001. Publishing house "DODEKA".
3. "Power electronics", B.Yu. Semenov, Moscow 2001
4. "Power semiconductor switches", P.A. Voronin, "DODEKA" 2001
5. Catalog of semi-automatic devices from NTE.
5. Reference materials from IR.
6. TOE, L.R. Neumann and P.L. Kalantarov, Part 2.
7. Welding and cutting of metals. D.L. Glizmanenko.
8. "Microcircuits for linear power supplies and their application", 2001. Publishing house "DODEKA".
9. "Theory and calculation of IVE transformers." Khnykov A.V. Moscow 2004

Homemade welding inverter next to the computer power supply:

The page was prepared based on the book “Welding inverter - it’s simple” by V.Yu. Negulyaev