Thank you, that is very generous. I’ll keep trying on my own until then. Knowing that it works on Qvadrans is a big help. The setup is almost identical aside from the use of 3PWM mode, and I have confirmed by two-channel oscilloscope that the mosfet gates are receiving inverted signals with proper timing. The only problem is the high side gate is not boosted to double voltage.
Today’s testing: Plucked the diodes from a Lepton with burnt mosfet, and put them on a little daughterboard which is connected to +V with a short wire though one of the bulk capacitor holes, and to the bootstrap capacitors using 34awg wire. No change in behavior.
