Matching reactive load

[Tony Williams]

So the problem might be the best way to drive a
1k//3nF load with at least 20V pk-pk and with (say)
better than 1uS rise and fall times.

[/QUOTE]

Put about 10.3uH between a 100kHz square wave generator and that
load...my simple brain is fascinated by the resulting waveform.

What generator resistance Tom?

With a zero ohm generator LTspice shows a 900KHz
resonant circuit being shock-excited every 5uS.
You can see that it is not a pickoff of the 9th
harmonic in the square wave because a similar
shock-excitation happens if it is resonated at
the 10th instead (8.44uH). There is also a
smoother transition at every incoming edge
because the 10th harmonic has a whole number of
resonant cycles every 5uS.

With a 50 ohm source resistance the Q drops so
low that there is a just slight peaking of the
100KHz square wave. Experimenting with LTspice
shows that 50 ohms and 2.5uH in series would
look to be a useful combination.... 50 ohms takes
the direct reactive load off the amplifier and the
2.5uh almost halves the risetime and reduces the
peak current requirement from 400mA to 270mA.

This is all based on a WAG'd load, but maybe a
pointer to the OP if he gets some real numbers.