Equivalent Models: Hybrid-pi vs. T-model for FETs

[Lunat1c]

Hi,

When it comes to the hybrid-pi model and the T-model for the AC equivalents of an FET, what's the exact difference?

For example I know that in a BJT circuit, the hybrid-pi includes the resistance $$r_\pi$$ in the base while the t-model includes the resistance $$r_e = \frac{r_\pi}{1+ \beta}$$ in the emitter instead.

In the case of an FET equivalent model, all I can notice is that in the hybrid-pi model, the gate is not connected to the source-drain junction, instead, it's represented as an open circuit given that the gate current is theoretically 0 and thus the FET can be said to have an infinite input impedance.

Also, for some reason, in the T-model there's a resistance included in the source which is equal to $$\frac{1}{g_m}$$, what does this represent? and why is it omitted in the hybrid-pi model?

I would really appreciate any help, I really need to know this because I have tests soon enough!

Thank you in advance.

 

[Valkarie]

The main difference between the hybrid-pi model and the T-model for the AC equivalents of an FET is that the hybrid-pi model does not include a resistance in the source, while the T-model does. In the hybrid-pi model, the gate is represented as an open circuit given its theoretically infinite input impedance. In the T-model, the resistance included in the source is equal to \frac{1}{g_m}, which represents the transconductance of the FET. This resistance is omitted in the hybrid-pi model since it does not affect the small-signal transfer characteristics of the device.