Adding a current limiter to my MOT welder

Last year I built myself a welder using microwave oven transformers; I used it for a variety of things, but it was far from finished. Recently I’ve done some work to finish off the main transformers and to add a current limiting feature.
Improved welder

The first thing to do was welding the main transformer cores together, this reduces vibration and allows more convenient mounting. Ideally I would have had access to a second welder, with which I could have welded the cores while they were not plugged in. Unfortunately I didn’t, so I had to settle for a slightly precarious method which consisted of gaffer taping wood over the windings and welding them using themselves. I sanded a patch of varnish off each core and clamped them one at a time in a vice, to which I had connected the ground clamp to complete the circuit.
welding the welder
I didn’t run into any problems caused by the cores being in use while being welded; they’re a bit ugly because I’m shit at welding, but work fine. Once they were welded together, it was a trivial matter to bolt them down to a board.

Previously it lacked any way to control the current; I frequently found that it was too powerful, It would burn 1/8″ rods and weld steel down to about 3mm, but anything smaller it would just blast holes and generally make an awful mess of. To sort this, I built a current using a third MOT, this time I’ll show how I stripped the transformer.

The first step is to cut the seam holding one side of the core together with a cutting disc in an angle grinder; once it has been ground out, a sharp upwards tap with a hammer will break any remaining weld or varnish and open the join up slightly.
opened up seam

It is now easy to pull open by hand, wobbling it a bit work hardens and breaks the other weld in no time.
opened up MOT

I then baked it in the oven at around 200°C for about half an hour, this was to help soften the varnish. After taking it back to the garage, I used a piece of wood and a mallet to alternately hit each end of the primary; after a few minutes it slipped fully off the core. Unfortunately in the process the paper wrapping remained stuck and pulled a few turns free of the main winding.
the slightly damaged primary
Luckily there was no damage to the insulation and I was able to carefully bend them back into position and wrap the whole thing in kapton tape to both hold them in place and protect the winding against future damage.

I was not so careful with the secondary (I’ve already got more lying around than I know what to do with), I walloped it out and in the process damaged the former and possibly the insulation in one or two places.
Mildly smashed secondary

Once I had cleaned any remaining lumps of varnish off the core with a file, I glued the winding securely into place with epoxy resin.
Winding epoxied to core

Without going into too many details, an inductor essentially acts like a resistor to AC currents (although an ideal one doesn’t dissipate any energy as heat); my plan was to put it in series with the primary of one of the transformers to limit the current going in and therefore reduce the current coming out. Using an MOT as an inductor is simple, the only tricky part is making it adjustable. Generally inductors consist of a coil and a core; changing the number of windings on the coil would have been tricky and would probably have resulted in bad connections and ruined insulation, but changing the core was fairly straight forward. The E and I sections placed together form a flux circuit; the aspect that matters here is that changing the cross section of this flux circuit will alter the total inductance.

It is easy to change this cross section by partially sliding the E piece off the top of the I piece; to support the core while this happens, I bent and welded a couple of bits of steel into guards and added some wood to support the E piece.
I piece with added supports

The core needed securely clamping together, but in such a way as to allow easy adjustment. I made a retaining bar to apply pressure from a piece of ~3mm steel (the handle I cut off an old gas bottle in another project) and added a pair of quick release bike axles to hold it down. They were a bit too long, so I used a die to extend their threads further up the rod then trimmed them down to size; I also reused the original nuts as hand adjustable lock nuts.
Current limiter retaining strap

I drilled holes and added spiked, threaded fittings to the bottom of the board then smeared some copper grease on the sliding surfaces and screwed everything into place.
Complete current limiter

I cleaned up the wiring with a couple of kettle lead sockets and added a switch to enable/disable the current limiter; Ideally it wouldn’t have had the middle “off” position, but it was the only 15A 250V switch I had to hand.
Sockets and switch

I installed some neat-ish wiring, leaving some slack in the current limiter wires to allow movement without any strain. I have not yet cut the board down to size as I intend to add a fan and/or carry handle at some point, but I have not planned out where these will attach.
Completed welder, showing the wiring

The full circuit diagram is as follows.
Circuit diagram
The two mains inputs are not connected directly together for several reasons, but if they must be plugged into the same phase for the welder to work properly. At full power they must go via separate 13A fuses, but when the current is limited to the lower end of the available range it is possible to use a single extension lead without blowing the fuse.

While I still don’t have a current clamp to measure the currents, it certainly drops it down to a more sensible range for smaller work. A partially completed version of it allowed me to weld a ~1mm steel box, and a setting slightly up from the lowest current allowed me to weld together some steel rods I had lying about that are a shade over 2mm diameter (that’s a 3.2mm welding rod for scale).
3 2mm rods welded together
Again, my abilities are a bit dubious, but the limiter clearly works (without it these rods would just have been destroyed).

28 thoughts on “Adding a current limiter to my MOT welder

  1. Pingback: Building an arc welder from microwave oven transformers | Products of Mike's Mind

  2. Pingback: Current limiter for a MOT welder

    1. Mike Post author

      I did consider this, however there are a few things that concerned me; I suspect the cores are already being partially driven into saturation (iron has got expensive recently, microwave manufacturers are stingy) which might complicate things, and as far as I can tell I’d be dissipating a notable amount of power into clamping diodes (unless I’ve misunderstood something about how the circuit would work).

      These things combined with the fact that I have loads of MOTs that I don’t know what to do with (the limiter was entirely built from things I had lying about) meant that I went with the mechanical method. I’d love to see a saturable core one to see how it compares, if you know of anyone who’s built one.

      1. Charlie

        Mike, I realize that your post regarding your MOT welder build is a bit dated, but I was wondering if you might be able to set me up with some 230v or 240v MOTs. I live in North America and all the MOTs are 115v on the primary. I’ve been searching ebay and the net for a bit of time without much of any success. I’m interested in a welder project but 115v primary just won’t do. If this is something you could do, please let me know the cost for 4 similar sized MOTs and post.

        Thanks, Charlie

        1. carz

          If you run your two primaries in series instead of parallel and run it on a 240v you will get only 120v to each primary. I did this with a welder and I am reading 120v per transformer.

    2. chalo hamisi

      Hi. Is the forum still running? Just completed the project minus the limiter. Works very well except the primaries get very hot after about a minute of welding. Any ideas how to rectify this. Will appreciate.

  3. Nicole N.

    Wouldn’t the wires connecting your variable inductor also present an additional inductance? While I do admire your careful attention to documenting and detailing your trials and tribulations; I can’t help wondering if your resulting welds are being impacted not from your ability to weld, rather the design of your welder. Inductors, including transformers, have energy storing properties which can create periodic reversals of the direction of energy flow. If you were to apply capacitors in parallel with the inductors you could slightly cancel out the reactive power that overheated your first transformers when you got carried away welding.
    Have you considered adding a full-wave bridge rectifier onto your output AC power so that you can weld using direct current and provide the choice to weld DCEN and DCEP to suit a broader range of materials? You can make a full-wave bridge rectifier with high current/high voltage Silicone Diodes (avoid Schottky diode versions.)

    1. Mike Post author

      The wires will add a tiny inductance, but the coil will have much much more (the iron core makes a huge difference). The reactive power is exactly what I’m using the inductor for- a perfect inductor would limit current without wasting any power to heat; in reality a variety of things cause losses (eddy currents, saturation and hysteresis mainly), but that can’t be fixed with capacitors.

      I would be interested in adding a rectifier; I had a look for diodes and couldn’t really see anything suitable, if you’ve seen some rated for hundreds of Amps at a reasonable price I’d love to know where.

  4. Michael Andrew Broughton

    if you connect a neon lamp between the hot leads of the 2 sockets, it’ll light up when they’re out of phase. and did you remove the shunts from the transformers? if you did, that’s probably what killed the first 2.

    1. Mike Post author

      The neon lamp idea is nice (and would be very useful for anyone copying this in a country with unpolarised plugs), but it will generally not cause any problems as domestic power here is usually single phase.

      I did indeed remove the shunts- I wanted the maximum possible current; overheating is an expected disadvantage unless duty cycle is kept low. I knew it might happen, but just got too carried away with my new toy.

  5. Josh

    Hey, Is a solution to this problem adding variable input power? Like a rotary light dimmer leading into the primary?

  6. Mike Post author

    There are a couple of main problems with using a light dimmer:

    Current- most light dimmers can handle enough current to light a few bulbs, perhaps an amp or two; this thing easily blows 13A fuses.

    Dimmers typically work by switching the current on and off rapidly (well, chopping bits out of the normal sine wave shape of the AC); this is fine for a resistive load like a bulb filament but will not play nicely with an inductive load like a transformer.

    1. Josh

      Apparently I have more reading to do! Would you mind expanding on “will not play nice”? –> Clearly I’m just uneducated in this regard, but it seemed to me that the end result would be the same on the secondary side, choppy flow, reducing the end wattage? Appreciate the quick reply and your willingness to chat! –> I’m working on a spot welder at the moment, but definitely consider myself a beginner.

      1. Anonymous

        Transformers are what are called an “inductive load”; instead of current being calculated at any point by V=IR (like a resistor), the current and voltage end up out of time (thinking of the current as having “momentum” is a fairly good analogy).

        Trying to abruptly start and stop a thing with lots of mass causes much higher forces or much smaller movements; the same is true of an inductive load. To be honest can’t remember all the details, but if you try it you will almost certainly get next to no power out or blow the dimmer.

        If you want to buy something off the shelf to “dim” a high power inductive load you can get something called a variac; it’s an adjustable transformer that allows you to gradually turn up the voltage while keeping the nice sinusoidal waveform. They tend to be pretty expensive and bulky, but sometimes something turns up on ebay. I would worry a little about using one for an arc welder as it will reduce the voltage and make it tricky to get an arc going; would be perfect for a spot welder though.

  7. Brian

    Looks like a fun weekend project. You might consider a Harbor Freight 90 amp mig welder for 90.00 then convert from ac like yours to dc (less splatter)The conversion can be done for about 30.00 and you get a awesome welder for around the house. I did a conversion the only difference was I added a bleed resistor to drain the capacitor in about a min

  8. Norman Burns

    Kitchen reciptles are normily wired with 3 wires giving the ability to have access to 240 volts without running 2 cords.
    By adding a variable wire wound resister to the coil of your current limiting transformer would allow you to vary your output current more easily

    1. Mike Post author

      I’m in England where we have 240V in every socket, we also have fuses in all our plugs. The highest plug fuse rating is 13A, which is less than the welder draws at full power hence the double plugs.

      Are you suggesting using a resistor on a secondary winding of the current limiting core to vary its inductance? That sounds like it’s going to dissipate a lot of heat for little advantage over mechanically adjusting the core, which only dissipates heat from hysteresis losses and winding resistance.

  9. Anonymous

    Could one use a variable speed controller such as the cheap 20 ones sold through harbor freight to control the volts sent to the transformers thus controlling output?

  10. ian f

    Wow. I’ve seen people around the net using 200$ variacs, high current scr circuits…you did it for free! I’m building one of these for mine…using a spring loaded linear track and a motor shaft /flywheel connected to a push arm with a vernier knob connected to the shaft…boom, precision amps control knob.

  11. Frank Sandlin

    To everyone looking for a simple circuit to control output of a mot welder .I think a mag amp circuit may work.what little Ive read on them you could actually control current and voltage on in and out of transformer by saturating core.may build up heat ,but a good cooling system should work.

  12. E Toft

    I just built a MOT welder. I use a Lutron ceiling fan speed controller that I bought at Lowes…it is designed for an inductive load…under $10.

    Works well so far. I also feed an old 12v wall wart power supply from the mains. The 12v feeds a fan from a pc power supply…one of the big 4″ ones…blowing across the transformer. I’ve attached aluminum heat sink fins to all sides of the xformer too. Still gets hot but I can do about 50 welds in 30 minutes…that is welding 10 guage high tensile steel wire.

    I am planning to use the 12v to add a trigger button and relay to “kick” the xformer on. Possibly will add an Arduino with a pressure sensor and automatic timer at some point too. By setting the pressure and time based on the materials being welded, it should turn out almost perfect welds.


Leave a Reply