Equations for input z are nonsense when dealing with rf. I just went through the experience of expecting 1 M ohm resistance from a Fet source follower, only to find out that with increasing f that 1 M ohm dropped way down to about 5000 ohms. Any comments?
Stray capacitance could be one source of the discrepancy. You've not indicated the circuit configuration and component types in any detail - just that it's a source follower. What frequency is of interest to you?
If you look at s-parameter plots as a function of frequency you see that nothing is constant. Input impedance is in fact a function of frequency. In particular, as you go up in frequency, things you can't see like inductance and stray capacitance are actually modifying the equivalent circuit.
That's right, Papa, thanks. And I'll be the first to point out that this effect is actually included in the literature. I had forgotten how the equations were modified by rf. Those 'tiny' device capacitances become significant as we increase the frequency, but the equations don't actually change; they're just modified by including device parameters and stray capacitance, even probe capacitance. I'm going through a time of transition from thinking audio frequencies to radio frequencies. And I'm beginning to see why rf engineers use power gain rather than voltage gain as a figure of merit for an amplifier.
Here's the circuit and the data. The graph of the curve looks like a low pass RC filter. It shows there is a total device input capacitance, stray capacitance, and probe capacitance of about 82 pF.