My resistance soldering gizmo is sufficiently handy that I keep promising to write it up here, but sufficiently weird that I keep not doing it. Here’s a first pass at rectifying that omission…

Back in 2007/8 I built a resistance soldering gizmo and wrote it up for my Circuit Cellar column (Feb / Apr / June 2008, issues 211, 213, 215). Those articles go into excruciating detail about transformer action, flux density, triac triggering with inductive loads, and how all the firmware works. If you want the gory details, go there to get the issues or there to get a CD.
What you’ll see here over the next few days is a quick overview of how I built the thing, along with some suggestions & color commentary.
The general idea is that you can get nearly all the spendy bits by harvesting parts from a kilowatt-class microwave oven. I used an ancient Sears ‘waver and I suspect older boxes will be better donors: more iron in the transformer, more robust semiconductors, bigger clearances. Use what you’ve got or can find along the side of the road.
That gives you the right half of the board in the picture. The transformer gets rewound for low voltage and very high current, the triac controls line current through the primary as usual, and the fuse does what it’s supposed to do.
Oh, yeah, the Vise Grip is providing a dead short to test the maximum current. It’s fine doing that; just makes the transformer buzz a bit. Remember, you don’t leave it on for hours at a time.
I wanted to show how triacs behave when they’re controlling highly inductive loads, which the (unloaded) transformer certainly is. So I built a controller around an Atmel AT89C2051 (the good bits of an 8051 stuffed in a 20-pin DIP) that gave complete source/sink control over the gate current in 1/8-cycle increments during six complete cycles. The left side of the picture therefore has some custom circuitry to make all that happen.
Like, for example, 2/3 duty cycle with maximum-voltage switching:

For a resistance soldering setup, you don’t want any of that. All you need is to turn the triac on and hold it on for a specific duration. Duty cycle control would be nice, but probably doesn’t make much difference. I’ll describe what I have, provide the source code, and you can hack it to make it do what you want…
Then I’ll show some electrodes and point to some projects I’ve soldered with the thing. It works fine and maybe you can get something useful out of it.

Onward…
Great, I’ve seen a number of spot welders on youtube based on this idea which looked dangerous. I’m looking forward to seeing more on a safer way of doing it.
Well, I’ll admit laying it out on that sheet of plywood is lethally stupid, but it made for a good picture for the magazine.
I think it’ll fit inside the carapace of a dead dehumidifier that will turn into a vacuum pump. That has the decided advantage of being plastic, so it won’t vibrate in the magnetic field.
So many projects, so little time…