My question is, does C(E) and r(E)(not showed) at Q1 have effect on the lower or upper -3dB frequency? I think there is, depending on the value of the C(E). Am I correct?

But the solution give for this question is:

Lower -3dB:
decide by : R(L1)//H(oe)+R3//R4 and C(c)

Upper -3dB:
decide by: R(L1)//H(oe)//R3//R4 and C(s)

 

Perhaps you are meant to assume CE is sufficiently large that it doesn't impact the -3dB cut-off point and plays a role well below the initial -3dB frequency transition.

 

Perhaps you are meant to assume CE is sufficiently large that it doesn't impact the -3dB cut-off point and plays a role well below the initial -3dB frequency transition.

So do you mean I am correct in theory, but the question here want me to assume C(E) is large enough.

Practically, in most situations, would C(E) play a role in the -3dB cut-off point, or usually not the case. Or it really depends on the specify circuit?

 

So do you mean I am correct in theory, but the question here want me to assume C(E) is large enough.

Yes that's what I mean.

Practically, in most situations, would C(E) play a role in the -3dB cut-off point, or usually not the case. Or it really depends on the specify circuit?

It depends on the relative time constants for the emitter and collector topologies. If one sees [say] 50Ω looking back into Q1 emitter terminal then what value of CE would ensure the emitter time constant is an order of magnitude greater that the derived collector time constant?