A staff, his second. Staff of Power came first. The primary is potted litz (400 × 0.1 mm in silicone gel) and the bus is doubled to 900 V. There is a segmented topload against eddy currents, a kilojoule of storage, a tungsten breakout point after the copper one kept melting, a 1 kW boost converter at 95 % and an aluminium body. The bridge is a 1200 V SiC full bridge run as APWM on Anders Mikkelsen's control scheme, so the card hangs under his. As far as the builder knows, it is the first full-bridge APWM. 3.5 m of arc off 140 mm of winding, 25 times the coil.
From the builder, 15 Oct 2026. The record is from a stationary test. The finished staff will be fully portable, and the ratio will drop a little for that. A startup oscillator makes it start at the upper pole. The driver sets a fixed dead time of about 100 ns. ZVS is not held through the first quarter of the ramp at switch-on, and switch-off is hard for most of it. One current transformer serves both feedback and the overcurrent detector, with about 100 ns of delay in simulation, and it only acts during ground strikes. The driver runs from a single 15 V supply, which bootstraps the high-side drivers and keeps the board small. The switches are Infineon SiC FETs with Kelvin source. The ones that died did so in an hour-long test at 6 kJ stored, 50 to 60 ms ramps and about 160 A, most likely from heat. Most other failures were litz touching something it should not have and arcing through. Daniël Eindhoven (“TeslaCommander”), an electronics graduate of The Hague (2010), works in power electronics professionally. The staff is near the end of its build, and he means to post the rest of the build details on his page soon. This card will be revised when he does.
Schematics and photos
photo: the builder, the staff lit in his hand 
photo: the record shot, two arcs across a garden at dusk 
photo: a single arc into the trees, daylight 
photo: arc to the ceiling of the workshop 
photo: into a tree at night 
render: the staff's head, segmented topload over the potted primary 
photo: the 60 mm full-bridge APWM power board, revision 2.1 
schematic: the power stage, revision 2.1, his own 
photo: the segmented topload and its tungsten breakout point 
photo: the primary capacitor board
Sources
This is the card from the driver tree, written from what the builder published. Send corrections through the forum.