Micro tig

Normally flow rate is set manually.

Right

not necessary

This is a fun project.

I once did the same thing and it worked, though it was not as feature rich as your project. I programmed the microcontroller in BASIC.

As Iggy said, the current ramp rate does not seem too useful and the gas flow rate is normally set at the flow meter with a needle valve. I very rarely change it.

One thing that you missed would be the AC "Balance" or ratio of electrode positive current to electrode negative current. The Duty Cycle would probably approximate that, but the fancy inverter machines allow setting them separately. Also, an external current control input is important, if you did not skip mentioning it because it was so obvious.

Something you might look at is that ST sells an eval kit for their IGBT modules and drivers that is a 100A max welding machine. It has been a long time since I looked at it, but it looked interesting when I saw it.

Sounds like a fun project, BobH

Unless you actually want the project, find a used Syncrowave 250, they go down to 5A output. Otherwise, lookup the specs on one and clone it other then the peak output.

I think that he meant this by saying "pulse share"

That's a useful thought, thanks.

Actually, no - but I don't really know what I'm talking about here.

I was thinking of when using AC, do you have a pulse on top of the AC, at presumably some higher frequency? Especially for very low overall currents, it might be useful, so that the higher current part stops the arc wandering?

Something like:

AC 100 Hz, pulse 1,000 Hz Forward: 6A for 1 ms 2A for 1 ms 6A for 1 ms 2A for 1 ms 6A for 1 ms 2A for 1 ms Reverse: 4A for 1 ms 2A for 1 ms 4A for 1 ms 2A for 1 ms

or maybe you don't need pulse on the reverse, only on the forward?

Because it's not a control on the panel, I guess. Yes, I will have a footpedal for off/on and current control.

I'll try and find that, sounds interesting.

I'm thinking of modifying a computer PSU with two independent 12v busses to give 24V at 30 A, adding a H-bridge for ac/dc and bang-bang current control, then adding another 24V at "up to" 1A DC on top to give 48V o/c, and HF on top of that.

I don't know whether I'll end up doing it that way, but it would be cheap.

Microprocessor will most likely be an Arduino nano plus.

Hope so, and thanks,

-- Peter Fairbrother

I hope that you are not doing this to save money.

Because you won't, at $10 per hour of your labor the cost will be astronomical.

I went through the same thing, it was fun and education, but it was a most expensive item I owned.

I am not sure on this. Maybe get the simple form working and when you like it, experiment, especially if it is a software change!

On your idea of using PC power supplies, which winding set is the output voltage controlled from? I was under the impression that it was the 5V, but that is just an assumption.

BobH

well, it needs a knob too - but I can repurpose that to do something else :)

I don't know. At least for the PSUs [1] I use for this sort of stuff, all 4 main outputs (+12v,+12V,+3,3v,+5V - but not the seperately generated -12V or +5v standby) are connected to the master controller chip.

I'm not sure exactly what that chip does - I have a datasheet somewhere

- but I don't care.

I would use the standby 5V to power the microcontroller and disconnect the main 3.3V and 5V supplies, then use the microcontroller to drive the optoisolator.

I also have an old-style, copper-wound mains, microwave oven transformer with a 20V 50A winding which I might use, but I think I'm going to save that for an EDM machine.

Useful for bench supplies, battery charging supplies, CNC stepper supplies, EDMs, electroforming supplies (the waveform for electroforming is about as involved as a TIG waveform BTW - however I used 555's, 557's and shift registers for that, and analogue current control (don't ask), not a microcontroller), electroplating supplies, and so on.

-- Peter Fairbrother