Uploaded April 2025 | Updated November 2026, 3 hours ago
Showed this one a couple times in the earlier stages and have just recently finished it up and put it in an enclosure. Was going to call it mini but ended up needing the full size of this enclosure. Phase Shift is probably the way to go but honestly at this kind of power level a front side buck isn't adding all that much to it. Thanks to Zak at QZW Labs I can see that it is viable to use a way smaller secondary and still produce nice long arcs. I'll probably try that for a QCW gun style build and hopefully another attempt at a low voltage push-pull using sic fets.
This is still using a full bridge of 60N65 247 package IGBT's and 2 UCC gate drivers w/GDT. Schmitt trigger feedback only.
The buck switch is a single 60N65 and is driven by an A3120.
I just stacked a couple 33x19x11 mm white/yellow toroids to wind the inductor.
The ZVS charging circuit is a real basic build using 55V fets and a hand wound transformer. The maximum charge output at about ~16.4VDC input is around 300VDC but it reaches about a max of 220VDC at 3 seconds charge time. This attempts to push a reasonable charge into the DC bus without working so hard as to kill the battery quickly.
DC bus is 4x CDE 1500uF 350V caps in parallel.
The ramp is fed to the A3120 using Finn Hammer's arduino code on an UNO powered by its own battery.
The enable input of the SSTC gate drivers is also connected to an arduino output through a standard optocoupler.
The end result is a somewhat dirty circuit that really needs more bus capacitance to increase the arc length but I don't want to add the extra weight. If I had to rebuild this again I would put the arduino inside the enclosure and set small trimmers and leave them for every adjustment aside from the BPS and On Time, which would be the only two I'd have adjustable from the outside. I would also use a charging circuit that drew little to no current to keep the caps fully charged and just leave it on all the time along with the controls. To utilize all my voltage regulators on the controls side I would just simplify it using a big 6s Lipo with a low voltage cutoff circuit on it. I didn't do that here because I was too lazy to rebuild the ZVS driver with higher voltage fets.
Showed this one a couple times in the earlier stages and have just recently finished it up and put it in an enclosure. Was going to call it mini but ended up needing the full size of this enclosure. Phase Shift is probably the way to go but honestly at this kind of power level a front side buck isn't adding all that much to it. Thanks to Zak at QZW Labs I can see that it is viable to use a way smaller secondary and still produce nice long arcs. I'll probably try that for a QCW gun style build and hopefully another attempt at a low voltage push-pull using sic fets.
This is still using a full bridge of 60N65 247 package IGBT's and 2 UCC gate drivers w/GDT. Schmitt trigger feedback only.
The buck switch is a single 60N65 and is driven by an A3120.
I just stacked a couple 33x19x11 mm white/yellow toroids to wind the inductor.
The ZVS charging circuit is a real basic build using 55V fets and a hand wound transformer. The maximum charge output at about ~16.4VDC input is around 300VDC but it reaches about a max of 220VDC at 3 seconds charge time. This attempts to push a reasonable charge into the DC bus without working so hard as to kill the battery quickly.
DC bus is 4x CDE 1500uF 350V caps in parallel.
The ramp is fed to the A3120 using Finn Hammer's arduino code on an UNO powered by its own battery.
The enable input of the SSTC gate drivers is also connected to an arduino output through a standard optocoupler.
The end result is a somewhat dirty circuit that really needs more bus capacitance to increase the arc length but I don't want to add the extra weight. If I had to rebuild this again I would put the arduino inside the enclosure and set small trimmers and leave them for every adjustment aside from the BPS and On Time, which would be the only two I'd have adjustable from the outside. I would also use a charging circuit that drew little to no current to keep the caps fully charged and just leave it on all the time along with the controls. To utilize all my voltage regulators on the controls side I would just simplify it using a big 6s Lipo with a low voltage cutoff circuit on it. I didn't do that here because I was too lazy to rebuild the ZVS driver with higher voltage fets.










