(Continued from the previous blog) As soon as I turned on the power, an astonishingly loud noise erupted, so I quickly switched it off. I inspected all the circuits, but there was  no mistake in the setup. I concluded that the loud noise was likely due to testing with the transformer’s core left unsealed, causing the  strong magnetic field to produce intense vibrations and sound. I realized that I should properly weld the cut section of the core before testing  again. However, welding it would make it difficult to address any issues if a defect was found, so I decided to try turning on the power switch once more. This time, I was prepared for the loud noise. I flipped the power switch on… and once again, a thunderous crash resounded, just as before.  But familiarity is a strange thing; once you’re mentally prepared, any action feels manageable. 

The voltmeter connected to the secondary coil showed the expected 4 volts, so I thought, “Alright! Success,” feeling satisfied. I then used  a DC welder (which I had modified to run on DC, also self-made) to weld the cut sections of the transformer core, securing them tightly in a vise.  Both welds were completed successfully. This homemade spot welder was made using a juice press base that was previously constructed from a log cut from a large oak tree in the garden. I initially built it to press grapefruit juice, as we have an abundance of grapefruits each year. However, it became unnecessary when my eldest daughter sent a high-performance juicer from the United States.

I welded the transformer core and secured both units to the base with screws. With this, 60% of the work was complete. Now, all that  remained was to secure the wiring with cable ties and, most importantly, to create the electrodes. After fixing the two transformers in place, I noticed that the secondary coil seemed a bit cramped. To improve insulation, I thought it best  to create some space between the coil and the core, so I inserted a wooden block against the wire and tapped it with a hammer to create a small gap. Now, it was time for the final check. Suppressing my excitement, I carefully double-checked the wiring and flipped the switch for the third  time. This time, a low humming sound filled the air—satisfying, but oddly, the voltmeter only showed 2 volts instead of the expected 4 volts.  Just as I thought, “That’s strange,” a large amount of smoke began rising from the transformer. It turned out later that the secondary coil had contacted the core in two spots, inadvertently creating a short circuit.

When I hammered the secondary coil, it damaged the insulation, causing the coil to contact the core and creating a complete short circuit.  Regaining focus, I cut the welded core open again, removed the secondary coil, and used a large burner to burn off the insulation tape and  paint. Reflecting on the inadequate insulation, I decided to cover the coil with heat shrink tubing, which I picked up at a nearby electronics store. It’s rare for things to go perfectly on the first try; clear mistakes and unexpected failures are part of the process. Ultimately, a positive outlook,  coupled with persistence and trial and error, is what leads to success.

This welder ultimately worked well and proved useful for repairing cut metal strips. Compared to repairs done with rivets, spot welding  allows for nearly instant repairs, making it a valuable tool. However, there is still room for improvement in terms of usability. In the image, the cut metal strip on the right has been repaired by securing it with rivets at two points marked with yellow circles. The strip  on the left has been spot-welded at four points, with the rust removed and repainted for reuse. Spot welding only takes 1–2 seconds, making it an  extremely efficient method.