So I came across this interesting MOSFET question and thought I'd give it a try. I seem to be stuck on a couple of parts however, hopefully someone can see something I'm not seeing. Here is the circuit:
Here's my attempt at a solution for it and where I've been getting stuck:
(a) In the low frequency regime, capacitor CL becomes an open circuit. Performing a small signal analysis on M1 results in the following gain equation: Av=(Vo/Vsig)=gm1*Req, where Req=(r01||r02) since there is no load resistor RL at the output. gm1 is calculable since ID1 can be determined from the current mirror formed by Mb and M2 and the provided value of ID3. But the question specifies that r01=r02=infinity, however. Would this mean that the low frequency gain is indeed infinity?
(b) We can consider the high-frequency regime to determine the 3dB freq and therefore must consider the internal capacitances of M1. Since there is no signal resistance (Rsig=0), you can use the amplifier output to determine the 3dB freq. In particular, fH=1/(2pi(CL+Cgd1)*Req), where CL includes the relevant capacitances from the other transistors. But the result is fH=0 since Req=(ro1||r02)=infinity.
To me, these answers don't seem correct, but I cannot figure out where I am going wrong. The fact that this circuit doesn't contain a load resistance and neglects channel length modulation is puzzling.