AC Arc Welder FAIL





Introduction: AC Arc Welder FAIL

I had the idea to make an arc welder from Tim Anderson's instructable "Build a Microwave Transformer Homemade Stick/Arc Welder". On his instructable he shows how to make a welder out of two microwave transformers.

I began working on it. After many hour of work and $80 spent it was finally finished. However, when I tried welding with it, it would not even arc. After half an hour of failed attempts I accidentally got my welding rod stuck to the metal I was welding. Before I even had a chance to remove it, one of the transformers began melting down.

What did I do wrong? Should I have used larger than 10 gauge wire for the new secondary coil? Should I have wound it more times? Fewer times? Do I have my transformers wired wrong?

If you can answer any of my questions please comment below. It would be greatly appreciated.



  • Epilog Challenge 9

    Epilog Challenge 9
  • First Time Author Contest 2018

    First Time Author Contest 2018
  • Sew Warm Contest 2018

    Sew Warm Contest 2018

We have a be nice policy.
Please be positive and constructive.




The most important things for the build, in no particular order:

1) Make sure you wind the wire exactly the same way in every transformer you use. If you wind the wire clockwise, starting from the bottom going up, then they better all be wound exactly that way.

2) Ground! All transformers should be mounted to the same metal base, and that metal base connected to the green (ground) wire on the electric cord.

3) Only use thin 3/32" or 1/16" rod. Standard US outlets can't supply enough current to weld with thicker rod. The amp requirements are listed on the back of the boxes that the rod comes in. Expect 30-40 amp at best.

4) Use a single electrical cord wired to both transformers in parallel.

5) Don't solder connections. At mains current, plain old lead solder will melt and create all kinds of hazards. Use physical crimps and connectors.

thanks for the help. I probably won't be finishing the project anytime soon, but I am curious to know what I did wrong. (I always want to learn, even from my mistakes) It really does make as much sense for me to buy a cheap one or borrow one for as little as I need to weld.

Well, what cooked the transformer was most likely that they were out of phase. All that means is that the voltage on one transformer wasn't changing in sync with the other transformer.

Basically, it's like two waves. If they are in phase, they merge together going in the same direction and add their power to each other. But the more out of phase they are, it's more like two waves crashing into each other. Most of the energy in a collision like that goes up as waste heat. So when the welding rod stuck, you put a heavy load on them and poof. Two big waves crash into each other and one ends up eating all the energy from the both of them.

Why that happened? There's a lot of things that can cause it. If the wire was wound differently on each transformer. Or the connections from the wall outlet aren't also matching. All it would take is reversing the hot and common (Black and white) wires going to one of the transformers and that would put it out of phase by 180 degrees. And with two electric cords, there's no guarantee that the 120 on one cord is going to be in phase with the other cord. If each one is plugged into an outlet on a different breaker then there's a good chance they will be out of phase. Just putting two transformers too close to each other can cause problems because their magnetic fields will interact too. So I guess it may be tough to say exactly what was wrong but probably swapped hot and common wires.

(Picture) The white and red wires are connected on the same terminal on each transformer. Can't see it in the picture so well, but all the transformers are wound the same way. And the whole thing is grounded via green wire to the metal chassis.


I don't think perfect phasing error with identical transformers would do it. If out of phase, the voltage would be zero. Two identical transformers connected out of phase and sorted would deliver no current to the load, and (other than losses) draw no current from the wall.
If the transformers were not identical enough, you could have some voltage, but too little voltage to strike and maintain the arc. Shorted electrodes, even with low voltage, would cause high currents to flow, which could overheat transformers and blow breakers.
When working properly, most of the output voltage from an arc weller is expressed across the arc, creating the heat to build the melt pool. A key skill of stock welding is to keep the arc length optimized for the energy dissipation needed, while not destroying anything.

Would you recommend using 240 instead? Or a 120 circuit with higher amperage? You'll have to post some pictures and wiring diagrams when you're done

For yours, if you wanted to take another crack at it, stick with 110, tear the burnt wiring off the transformer. Use the same 10 gauge wire and wind it again making sure it's wound the same way.

When you connect the windings together, make sure to connect them together in series... i.e. the windings have a beginning and ending, connect the ending bit of wire from one transformer to the beginning bit of wire on the other transformer. So you'd have the first bit of wire (That you started the winding) connected to your clamp, and then the last bit of wire on the 2nd transformer connected to the rod holder.

After that, only use one power cord and make sure the wires connect the same way on both the transformers. If they're facing the same direction, the white plugs should both be on one side, and the black plugs on the other.

If you get 40vac or so once it's plugged in, you should be ready to go for 1/16" or so welding rods.

In retrospect, I'd just get a cheapo harbor freight special welder rather than mess around with using these transformers.

On my work in progress, it trips the breakers constantly on a regular 15 amp breaker. Hooked it up to 30 amp. Doesn't trip breakers anymore, but it is now more of a plasma cutter than a welder. The only way it works at all on 15 amp, is if you use only two transformers, they both need to be putting out as close to 20v as possible each (so 40vac when in series), and then it will only work with 1/16" or so rod.

On mine, I picked up a big IGBT and a couple heat sinked bridge rectifiers and plan on fixing it up nicely with controllable amperage , steady dc rather than AC, etc. But by the time I get it sorted out, it wouldn't have been any cheaper or easier than buying a machine. And I came about two inches away from killing myself with the thing... sooo, I don't think the time, and risk to life and limb is worth saving some money and getting an unregulated, dangerous welder.


there's a DIY Jacob's ladder video on Youtube that shows how microwave transformers can overheat when abused.

it is essentially the same principle. Your rods have to also be thin enough to melt. Thicker rods in my limited experience required more power, and the rods you used might have pulled too hard on the transformers. I'm no expert. Hope this helps.

further their is a duty rating on welders. the box has no ventilation. was the box real hot? i had a harbor freight welder with a 9% duty cycle. a couple of 9 inch 110v fans i put on the transformer to increase the duty cycle.

um no facts. on small low power welders it sometimes takes a minute to get them to arc. lots of sticking. by the rod sticking you had power. a fuse would have saved the transformers. without a volt amp measurements the rest is conjecture. if you game rebuild and be somewhat conservative in your tests. if the turns are backwards, or the two transformers are wired backwards. could mean a world of difference in functionality.