Cheap Banggood welder to Cold Welder conversion

[tek2222]

I want to mod a Cheap orange Banggood DC Welder to control the welding time and current accurately from a microcontroller. Also i want to make it possible to create a spark from the microcontroller.

I can enable and disable the welder via the shutdown pin from a raspberry currently , so it is easier to create and tweak the necessary state machine, the plan is to start welding with an input from the main Button of the Tig torch. Then the raspberry would enable the welder pwm by pullling the shutdown pin low. A short time later the raspberry will enable a HF coil with a transistor and create a spark from the gun tip to the workpiece, lighting the arc. A fixed time later the raspberry will again shut down the PWM of the welder to end the arc.

One question that i have would be what would be a good sensor to sense the arc as being lit, otherwise it might be difficult to predict when exactly the arc lights up. I could imagine to measure the voltage on the Welder electrodes, but im worried that High voltage spikes might destroy the raspberry, another option would be to use a loop style current sensor , but im not sure how quickly it could measure a current that rises when the arc lights up.


I'll post updates and photos with progress,

I'm new to this forum and I wanted to post here to see if someone is interested in this project.

[tek2222]

making good progress...

The image shows the raspberry,
currently i tested

2 seconds welder off
0.1 seconds welder on (lowest setting)

[tek2222]

This is what happens to a piece of filler wire that i dragged across a CDRom housing.
The two smaller spots come from touching the surface while the welder was on 0.1 seconds
The third larger spot created a larger arc and actually made a small weld.

going to add HF arc next.

[tek2222]

Got the High Voltage Arc to reliably start with a Relay.
The LED indicates "HV Arc start"

[tek2222]

First welds :
The smaller one is 0.1 the bigger ones 0.2 sec

The images show the backside and front side of a CDRom drive metal housing.
backside shows more oxidation because it does not have argon.

The gif shows the middle (0.2s) and the smaller weld of the 3 to the right (0.1s)
Welder was set to 45 amps, but these welder currents are off.
I wonder if it is better to use shorter pulses and more current, or less current and
longer pulses. I guess with a shorter pulse there is less chance of the process getting unstable ,
spraying sparks and such.

[tek2222]

Here some manual welds,
the hole is where i forgot to turn on Argon Gas.
there are some different current settings used, but all at the same duration.

[tek2222]

So after having fun with welding some stuff like that  and sucessfully reparing smaller household items with that method (dishwasher rails that were rusted through and a camera holder), i tested some more with different settings, and when i set the amperage of the welder to full (whatever that is) i think it blew the input IGBT and damaged some more on the input stage, now blowing the fuse when switched on.

I posted the schematic below, I would like to get some help identifying problems with this design,

I can imagine two main problems:

- switching on the welder quickly with the shutdown PWM pin and then welding could maybe not what the welder was designed for, possibly a bigger capacitor would be needed ?

- the HV spark could have blown stuff by getting into the welder ....
Do i need a more beefy welder, or what are the No-No's that i could solve with a different setup ?