<xentrac>
the left-hand part of the circuit is generating PWM, the lower right uses two capacitors and a transformer to reactively filter the PWM signal
<xentrac>
but I wanted to do the filtering purely reactively (or as close to purely reactively as possible) in order to maximize the power efficiency
<xentrac>
eyeballing the plots, it does seem like it achieves better filtering than *just* the RLC low-pass filter corresponding to the relevant part of the transformer and the output cap
<xentrac>
but I feel like I don't know enough about filter design to know if a π LC filter or something would work better with the same number of reactive components and no inverter
<xentrac>
what do you think?
<xentrac>
(apologies for using Falstad's simulator here; I don't know how to use SPICE or WR-SPICE yet)