MOT Spot Welder

[admin]

This thread will detail the Microwave Oven Transformer (MOT) Spot Welder Build.

A MOT can instead of converting to high voltage be re-purposed to deliver high Current (400-1000A) @ low voltage (1 - 3V). This high current can be used to melt metal and form the basis of a spot welder. 

The build is still in progress and will be updated as progress is made:

The original plan is as follows:
   

Firstly we start of with a nice big Mother of a Transformer (MoT):
The primary loop resistance measured 1.6Ohms with the secondary loop measuring 75.9Ohms 
       

In order to re-purpose the MOT we need to remove the "secondary" coil which can be identified by the thin wire it is made off. This coil is then replaced with a thick wire of 1-3 loops.

The transformer with secondary removed:
The availible space with the Shunt removed 19mm wide by 39 height.
   

And that is where I am now.

A word of caution, I wanted to proceed using wire from a vehicle "Jumper Cable". I actually found a nice thick 16mm (diameter) jumper cable at MIDAS rated 1000A. However please be cautioned when I examined the actual wire core was only 3mm or 4mm in diameter with the plastic insulation being the majority of the cable. I do not believe this cable is suitable for this purpose and will continue my search for actual AWG 2 to 4 cable. 

In the meantime I will continue work on a Triac based driver circuit.

[chemcool]

A word of caution, I wanted to proceed using wire from a vehicle "Jumper Cable". I actually found a nice thick 16mm (diameter) jumper cable at MIDAS rated 1000A. However please be cautioned when I examined the actual wire core was only 3mm or 4mm in diameter with the plastic insulation being the majority of the cable.

I was and is also concerned, using that same cables in my solar project at home.

[Stevo]

For Your Triac, look for a Snubberless one, then you don't need to create a snubber. MOC3021 Triac Driver and for the Zero Cross detector circuit, I recommend using the AC Opto Couples http://za.rs-online.com/web/p/optocouplers/6912085/ or http://za.rs-online.com/web/p/optocouplers/8130733/

You using a Microwave Transformer there ?

Here is a quick and easy spot welder without the whole triac setup.

https://www.youtube.com/watch?v=vrlvqib94xQ

[Gordon]

For Your Triac, look for a Snubberless one, then you don't need to create a snubber. MOC3021 Triac Driver and for the Zero Cross detector circuit, I recommend using the AC Opto Couples http://za.rs-online.com/web/p/optocouplers/6912085/ or http://za.rs-online.com/web/p/optocouplers/8130733/

You using a Microwave Transformer there ?

Here is a quick and easy spot welder without the whole triac setup.

https://www.youtube.com/watch?v=vrlvqib94xQ

The original reason for starting this build is to do Battery Pack repairs and spotwelding of small batteries as well as bigger tasks. Without a timer or some method of controlling the power/length of weld it could be a fingerless experiment on batteries  from what we have investigated it appears a dual timed pulse is safest for batteries and will also with longer pulses work on heavier jobs.

Even thicker materials would benefit from timed welds which most of store bought devices do to some degree.

---

http://www.mantech.co.za/ProductInfo.aspx?Item=82M0942 Would the IC 555 work? Seems a simple circuit without the need for a microprocessor.

[admin]

@Chemcool, Its scary how I almost fell for it,  The cable really looks the part. I have got it on good authority that "LIDO" in Randburg is the company I need to go see for proper cable. Will check them out tomorrow.

@Stevo, Yes Microwave Transformer is being used. Will look into the Triac design next.

@Gordon, My use case will probably be more sheet metal spot welding and less batteries, this is why I'm trying to get hold of the correct cable (to handle the higher current). The dual pulse timer is really important, however I believe the triple 5 timer is too primitive for this application, its very good if you wanted only a simple single pulse, however to get two configurable length pulses in sequence could prove challenging. I tending towards using an Arduino Uno for this project. I see Org Jansen made some serious progress on his build. What are you looking to build next? the timer or the driver? Maybe worthwhile trying to put together Orgs driver in the meantime and testing it out?

Also, I would love to start seeing your guys build on the forum. 

[Gordon]

@Chemcool, Its scary how I almost fell for it,  The cable really looks the part. I have got it on good authority that "LIDO" in Randburg is the company I need to go see for proper cable. Will check them out tomorrow.

@Stevo, Yes Microwave Transformer is being used. Will look into the Triac design next.

@Gordon, My use case will probably be more sheet metal spot welding and less batteries, this is why I'm trying to get hold of the correct cable (to handle the higher current). The dual pulse timer is really important, however I believe the triple 5 timer is too primitive for this application, its very good if you wanted only a simple single pulse, however to get two configurable length pulses in sequence could prove challenging. I tending towards using an Arduino Uno for this project. I see Org Jansen made some serious progress on his build. What are you looking to build next? the timer or the driver? Maybe worthwhile trying to put together Orgs driver in the meantime and testing it out?

Also, I would love to start seeing your guys build on the forum. 

I started the thread in the house4hack. Just didn't check before I went and subscribed with the wrong gmail account I can go past Lido on Thursday as I had planned to stop in at the welding shop to check what cable they had and also for tips. I have a mig welder so I have loads of tips from that but want to see what the right stuff costs as I'm sure they will also last a lot longer. Will feedback from both places on what I find. I'm meeting with Org tomorrow night to check his out. Also I read up a bit and it seems my 266amps is not enough even for battery tabs. Need the 400 and for 2mm 800amps and it jumps up much higher. The purchasable spotwelders are 2000amps + and "apparently" they only do 5mm. Not sure about all the info but once we have it working switching cable is actually the easiest and probably cheapest part (next to the MOT itself) of the whole project. Sheet metal is also something I want it to do so I can make cases and boxes.

[admin]

The Driver Circuit:

Okay so I'm happy with the following driver circuit design. I started with the datasheet for the "TRIAC OPTOISOLATOR" MOC30XX. 

We can skip through to the "typical circuit" on page 7 as this summarizes everything (seriously):

   

Firstly notice something very important, the MOC303x Range is recommended for 115V AC operation as can be seen in the Figure 12 circuit. For 240V AC (like we have here) the MOC304x Range is recommended (as is withstands a higher 400 volt peak). The MOC3043 is the specific part that I chose as it has the lowest led drive current i.e. 5mA to get it going, ideal for a microprocessor to drive directly. Finally as it is mentioned in the last line above the 39 Ohm resistor and 0.01 Cap is only required if the selected Triac needs subbing. 

The Triac ( 25A_TRIAC ) selected is a snubberless (does not require snubber circuitry) TRIAC and as such we should be able to remove the snubber mentioned above.

The final Circuit that I will build for the "Driver" will thus look as follows:
   

Will try and get the parts to build and test. I like to keep things simple, please let me know if I have left out anything.

[admin]

So I managed to pop by Lido,

The cable that may suit the spot welder is sold as "Welding Cable", They sell them by the cross sectional area size so you would ask for 35mm,50mm,70mm sizes.

35mm^2 (R45/m):
   
This cable is 12.4mm in diameter.

50mm^2 (R82/m):
This was out of stock however they say they should arrive as soon as tomorrow.

70mm^2 (R95/m):
   
This cable is 16.85mm in diameter.

I grabbed 2m of the 70mm wire as this should fit 2 loops nicely in our transformer gap. I would suggest the 50mm if you want 3 loops. 

And finally I grabbed a few "Lugs":
   

[admin]

After a lot of messing around with the driver circuit (and pulling out whats left of my hair) I must admit that I made a mistake, the whole zero crossing Triac optocoupler (MOC30xx) was a bad idea for two reasons. 
1. I could not get the "typical" circuit to work. 
2. For our application, Zero crossing is better implemented on the microcontroller side, this means that we can start our "timed" pulse exactly from the 0 point after the trigger is pressed. If we had the zero crossing within the optocoupler our micro's "ON" signal may be delayed until the zero crossing occurs.

To cut a long story short, Org was on the right track and using a "random phase" optocoupler is the correct way to go (MOC302x). Baie Dankie Mineer Org for sharing your circuit.

On the wire side, I used 2 turns of the thicker 70mm^2 wire (fit perfectly snug) and achieved a 2.11V output and 850Amps. I managed to make a thick nail red hot as well as do some very good sheet metal spot welding (1mm).

   
   

I have the driver circuit on breadboard and will transfer to vero shortly.

[admin]

Figured out the issue with my zero crossing optocoupler circuit above... A Triac's T1,T2 terminals are NOT INTERCHANGEABLE. 

New circuit used:
   

Completed driver:
   
   

Now to the Arduino controller.