Do-it-yourself spot welding on a microcontroller. Resistance welding from a microwave and a homemade timer on PIC

The power electrical circuit of the spotter has long passed the stages of development and experimentation and is used for straightening cars in a variety of ways. After gaining experience with the device, the question of automatic control of the device’s operating modes with more precise adjustments and the necessary protections arose. A spotter with a mode and a spotter as a welding machine for working with an electrode must have different pulse durations and power. The welding point may turn out to be weak or too strong, which will create additional difficulties when repairing the car.

Photo 1. The spotter is indispensable when carrying out automotive body work.

The main parameters that require precise adjustment for a high-quality work result are the pulse power and its duration. The proposed scheme will allow you to select and save parameter settings both in the welding machine mode and when doing spot welding.

The circuit is assembled on three boards and consists of two functional parts:

1. The board on which the power supply is located. The appearance can be seen in photo 1.
2. Two boards, one of which contains a controller and the second with switch buttons and a four-digit indicator.

Power supply and its circuit

The power supply diagram is shown in Fig. 1. Conventionally, it can be divided into three components:

• power circuit of the primary winding of the step-down transformer;
• a step-down transformer;
• secondary winding with diode bridge and voltage stabilizer.

A surge filter, usually used in switching power supplies, is installed in the primary winding circuit of the transformer. Here it is used to protect the controller chip from impulses created in the mains voltage during operation of the spotter.

Any transformer with a voltage of 220 V/24 V can be used when operating from a 220 V network. When operating from a 380 V network, you need to use an appropriate transformer and a surge filter.

A diode bridge with smoothing capacitors and a voltage stabilizer on the LM2574 chip are connected to the secondary winding. From the output of the microcircuit, a nominal voltage of 5 V is supplied to the output connector X1 through an LC filter chain to eliminate high-frequency interference. The connecting lines marked with a dotted line should be of a minimum length and located as close as possible to the second leg of the IC1 chip.

Figure 1. Power supply diagram.

The voltage at terminal 1 of connector X1 is used by the controller to determine the zero level.

The voltage from terminal 7 of connector X1 is used to start the controller at a positive half-wave of the mains voltage.

A self-made circuit, if there are no errors in the assembly, starts working without additional settings. The presence of a voltage of 5 V will control LED1.

Starter K1 is designed to connect mains voltage when switch S1 is closed.

Instead, you can use a circuit breaker with protection of the required rating or connect the voltage directly, if there are fuses in the supply network.

Return to contents

Control of power thyristor spot welding spotter

Photo 2. External view of the control unit board with controller.

To control a power thyristor or triac, the MOS3052 microcircuit is used. This series of microcircuits is specialized for use in devices of this type and when replaced with analogues. In this case, it is necessary to carefully evaluate the technical characteristics of the proposed option.

When powering the circuit from a mains voltage of 380 V, it is necessary to use a triac of type VTA40 - 800v, ​​respectively, the operating voltage of capacitor C11 is 630 V, protective varistors R14 and R15 of type 20D241. To install a triac, you need to use a radiator. The design of the element is safe and has no connection to the heat sink. To control the temperature, it is advisable to install a thermostat with a contact opening temperature of 60-80°C on the radiator. A power transformer can be equipped with similar control. An alarm signal from thermostats can be connected to the controller to stop operation when the temperature exceeds the permissible temperature, with the corresponding signal displayed on the indicators.

For high-power spotters, we can recommend another version of the thyristor control circuit. It uses 70TPS12 type thyristors, which are controlled by MOS3052 optocouplers. Thyristors of this type have an electrical connection to heat sinks and must be installed on separate radiators or with dielectric spacers.

Return to contents

Control circuit with indicator block for spot welding spotter

Figure 2. Diagram of the control unit for the spotter.

The appearance of the control unit board with the controller is shown in photo 2.

The photo shows the appearance of the indicator block with control buttons without a decorative panel. The indicator panel with buttons and installed decorative panel is shown in another photo 3.

The control circuit has a minimum of auxiliary elements. All processes are controlled by an AtMega 16 microcontroller installed in the DIP version. The element from the manufacturer Atmel has a low cost and a large number of pins. The controller device allows the use of input and output signals on any legs of the microcircuit, so the board is as simplified as possible. In addition to configuration capabilities, the controller is equipped with high-capacity RAM and non-volatile memory, etc. In the spotter control circuit, its capabilities are used by approximately 20%.

Return to contents

Brief description of the operation of spot welding spotter

The schematic diagram of the control unit is shown in the figure (Fig. 2). When the supply voltage is supplied, the data stored in non-volatile memory for the first button is loaded. The indicator displays the information provided by the controller. In parallel with the output of information, the state of the buttons is monitored; when a triggered button is detected, the corresponding subroutine is launched. The information on the board is updated in connection with the new request.

Each time the button contacts are activated, a sound signal is heard; its absence means that the controller is malfunctioning or frozen.

Photo 3. Spotter indicator panel.

Using the buttons, you can select the desired operating mode and set the desired pulse parameters. The selected mode can be saved in memory for later use.

In the "Operation" mode, the controller operates as follows:

1. The indicators turn off, the controller monitors the voltage level at the AIN1 contact.
2. When the voltage drops to zero, the counter starts with a set pause period.
3. At the end of the countdown, a command is issued to the thyristor (triac) control chip. The process is repeated at each cycle of the mains voltage to use only the positive half of the cycle. This improvement avoids the magnetic saturation mode of iron.

Mains voltage control occurs along the chain from the power supply, through the X-1 connector contact to the SIN controller contact. Elements VR2 and Q2 correct the signal shape. The voltage to open the triac is supplied to connector X3, pins 1 and 2.

I stripped the ends and tried on the lugs - they dangled freely on the wire.

There is definitely something wrong here and I really wanted to figure it out.
I measured several copper wires more precisely with a Soviet micrometer - an average of 0.365 mm came out
And I settled down to count them more comfortably...


Counted 433 pieces
Through simple mathematical calculations, the actual cable cross-section was determined to be 45 sq mm.
It won't be enough, it won't be enough!
How can this be, since you saw the tag on the cable with your own eyes? And this is how they deceive gullible buyers. In many specialized stores, when purchasing cables and wires, sellers even ask whether the cross-section is required (according to GOST) or underestimated (according to TU). Moreover, even the cross-section of the wire according to GOST is also underestimated - it has been checked several times. Stranded wires are underestimated more than solid wires, because... It is difficult to check their real cross-section. In this case, the PuGV 1x50 wire with an already reduced cross-section was signed as PuGV 1x70.

So, the actual cross-section of the wire is 45 sq. mm, which is still not enough for such a transformer. It was not possible to quickly find a meter of wire with a real cross-section of 70 sq. mm, so I will test it on what I have (later I may redo it). I also decided not to change the tips, because... I will not press them, but solder them.

The process of soldering such thick wires at home is not a trivial task, so I will describe in a little more detail how this is done.
Take the most powerful soldering iron available and put it aside - you won’t need it :)
However, you can try it with a soldering iron.

An assistant to increase the number of hands is highly desirable. Unfortunately, no one helped me, so the process was not as convenient as it could have been and, of course, I didn’t take photos during the process - my hands were very busy, I’ll have to describe it in words :)
Soldering was carried out with a Chinese gas torch of medium power (1 kW stated)


The soldering location was chosen in accordance with fire safety requirements, away from flammable materials.
I stripped the ends of the wires with a margin so that the insulation around the tips did not burn too much.


I first put on heat-shrinkable tubes to later isolate the soldering areas.


The transformer was lifted and secured higher, the wires were bent down vertically - in this position they should be soldered. I wet the wire with flux, put on the tip, and bend the wires sticking out in the control hole so that the tip stays on the wire. A wire with a fair cross-section will not fall off anyway, since it is inserted into the tip with considerable force.
I heat the tip together with the wire to a temperature of approximately 220-230 degrees (in about 1 minute) and insert POS61 solder wire into the gap, which melts and fills all the free space. This takes a couple more minutes, while I continue to slightly warm up the tip. As soon as solder appears in the control hole, I stop soldering and slowly cool everything. The second wire was soldered in the same way

Next, I pulled the tubes to the tips and pressed them with a hairdryer in two layers.

To transmit maximum power, the power wires should not be too long, but very short wires make the welding process difficult. My length turned out to be 35cm, it could have been made a little shorter.


For convenient starting, the button was attached to the power cable next to the tip (visible in the photo)

To weld batteries, I cut out copper electrodes from 2mm plates


And bolted it into place

The display is very fragile, it is advisable to protect it better during installation, I didn’t do this, maybe I’ll redo it later.

The first thing I checked was the nickel tape.


Width 6mm, thickness 0.14mm and length 500mm
The cross-section is 0.84 sq mm, the measured resistance is 0.051 Ohm, the specific conductivity is 0.086 Ohm*mm2/m, which corresponds to nickel.
The conductivity of nickel is 5 times less than copper, which, together with the small cross-section of this tape, does not allow it to be used for assembling batteries for powerful power tools. For such assemblies, you need to use 10x0.2mm tape with a cross-section of 2 sq. mm or even solder the batteries with a copper conductor of 1 sq. mm or more (which is what I usually do).

Testing the welding controller and the welder itself
Adjustment limits:
Pulse duration 10-200ms, default 40ms
Number of pulses 1-10, default 2
Pulse shift relative to zero: 0-10ms, default 2ms
The pause between pulses is equal to the pulse duration
The operating mode is not saved after a power failure, but you can overwrite the default settings by holding the encoder button for 10 seconds.
There are no presets or profiles, but due to the small number of settings they are not needed

After pressing the start button, the indicator says WELDING (welding), a loud warning signal sounds 3 times, then the welding itself begins and at the end the welding end signal sounds 2 times.
The green LED on the board indicates ready mode. It goes out during the welding process.

As in any business, to obtain a normal result you need skill and training. Resistance welding has its own area of ​​application and this must be taken into account.
Do not try to immediately cook new expensive batteries, because... There's too much chance of ruining them. Train on old or faulty batteries to select the shape of the electrodes, clamping force and welding modes.
A little theory.
The specific power at the contact point is (I x U x T) / S
T (pulse duration) can be selected in the controller parameters
U (voltage at the contact point) depends on the transformer and the passing current
I (current) depends on the transformer, electrodes, clamping force at the point of contact
S (contact area) depends on the shape of the electrodes and their pressing force
As you can see, there are quite a lot of influencing parameters, so we have to select them.
For example, you should not try to make blunt electrodes or put too much pressure on them, because... Despite the high current, the voltage at the contact point will be very small and naturally there will be no normal heating. You should also not space the welding points too far apart, because the current will not be able to reach the required value due to the high resistance between the contacts.

Due to the synchronization of pulses with the network, the repeatability of weld points is quite high. All tests are tied to a specific device - the results may naturally differ on another device.

Welding a battery in different modes (from left to right)
1/10 1/20 1/40 2/40 2/60
The first is the number of pulses, then the pulse duration


The optimal value is 1/40.

Welding AAA battery, mode 2/20

Welding paper clips

Below is shown how not to cook batteries :)


Dull electrodes and high clamping force.


In this case, the power is released not at the point of contact, but in the wire itself - naturally, nothing is welded and the plate easily flies off

Welding a battery at one point with blunt electrodes (one electrode on the battery, the second on the plate)
There are 2 points due to welding 2 times


It’s too easy to burn out the battery, and the welding doesn’t last


If you really need normal welding at one point, make one electrode blunt - and press it harder against the battery so that heat does not generate in this place.

Burnout in 2/60 mode


Overburning may compromise the seal of the battery, which is unacceptable.

Welding in the wrong place on the side surface


Left - 1/40ms mode, right 2/60ms (overburn)
There is no protective gasket on the inside side surface and welding can damage the battery roll.

During the welding process, the batteries, transformer and triac do not have time to heat up, but if a more powerful transformer is used and welding is intensive, forced cooling may be necessary

Wishes to the manufacturer.
1. Add a welding mode without a preparatory delay (for pedal control)
2. Add a welding mode by holding the button (for welding massive elements with long exposure times)
3. Provide the ability to turn off the loud squeak (at least with a jumper)
4. On the board, change the rotation of contacts to the display (so that they match)
5. Make the pulse duration setting scale two-zone, for example from 10 to 100ms - in 1ms steps, over 100ms - in 10ms steps

Conclusion: the controller performed well and can be recommended for use

Fluffy refused the photo shoot - the suspicious piece of iron with thick wires scares him.

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