Well,

Still working on finding resonance.

When using a function generator in conjunction with a spectrum analyzer some interesting things happen. The spectrum analyzer shows that as the applied frequency changes the peaks also change. To me this suggests the capacitance changes drastically with frequency (and it's not linear). That is one reason why it's so hard to find resonance on this circuit.

Another interesting thing is that the circuit acts as a filter. At times the square waves turn into sine waves, or amplitude modulation. And at some frequencies the square wave is multiple times lower in frequency than the applied square wave. For instance at 20.5kHz applied the wave across the cap is 37Hz.

[attachment=1987]

Above: Notice the applied frequency of 20.5kHz and the width of the pulse (27mS) is 37Hz.

Not sure exactly what's going on here. But the capacitor is definately a non-linear device. More to come.

Still working on finding resonance.

When using a function generator in conjunction with a spectrum analyzer some interesting things happen. The spectrum analyzer shows that as the applied frequency changes the peaks also change. To me this suggests the capacitance changes drastically with frequency (and it's not linear). That is one reason why it's so hard to find resonance on this circuit.

Another interesting thing is that the circuit acts as a filter. At times the square waves turn into sine waves, or amplitude modulation. And at some frequencies the square wave is multiple times lower in frequency than the applied square wave. For instance at 20.5kHz applied the wave across the cap is 37Hz.

[attachment=1987]

Above: Notice the applied frequency of 20.5kHz and the width of the pulse (27mS) is 37Hz.

Not sure exactly what's going on here. But the capacitor is definately a non-linear device. More to come.