Questions about oscillators （in Razavi book&#

[wangkes9]

1. Figure1 ;Why the circuit must only oscillate at 180 degree .
IF H(jw)=10 ,Angle (jW) =-179 and H=0.5 ,Angle (jw)=-180 , then the cuicuits will oscillates at the frequence W which phase shift is -179;

2. Figure 2, in open loop ; the Wosci = 3^0.5 W0.
In close loop, the wosci = A0*3^0.5/2 W0 (seen in cos(…t))
Which is the right value?


Thanks!

 

[erikl]

Re: Questions about oscillators （in Razavi book&a

2. Figure 2, in open loop ; the Wosci = 3^0.5 W0.

Closed loop of course. No oscillation in open loop!

In close loop, the wosci = A0*3^0.5/2 W0 (seen in cos(…t))
Which is the right value?

Equ. 14.16 is valid for A0 ≥ 2 ; so it's the same!
Because of saturation/limitation effect, A0 will never be » 2 , i.e. A0 ≈ 2 .

 

[wangkes9]

Re: Questions about oscillators （in Razavi book&a

Thanks.

Any one can help the first question?

 

[LvW]

Re: Questions about oscillators （in Razavi book&a

Thanks.

Any one can help the first question?

The circuit consists of 3 inverting stages - thus, the whole circuit is inverting (180 deg phase shift) .
When each stage contributes additionally 60 deg. because of "parasitic" phase shift, you arrive at 360 deg. which fulfilles the oscillation criterion.

 

[erikl]

Re: Questions about oscillators （in Razavi book&a

Thanks.
Any one can help the first question?

For me it's not quite clear how you mean this:

IF H(jw)=10 ,Angle (jW) =-179 and H=0.5 ,Angle (jw)=-180

If you think of a 2-stage-amplifier, this would mean an overall H(jω)=5 and an overall AngleH(jω)=-359° . There will always be a (slightly different) frequency ω0=ω+ωx , for which both Barkhausen criteria are satisfied, i.e. the oscillation frequency ω0 will always occur at an overall phase shift of ±180° (negative feedback) or ±360° (positive feedback), s. also /Razavi/ Fig. 14.3 on p. 484 and its explanation.

However there could be an exemption: If the overall gain |H(jω0)| at the 2nd Barkhausen criterion (|AngleH(jω0)|=180°) were less than 1, but |H(jω)|≥1 at an |AngleH(jω)|=179° (for ω<ω0), then I'm fairly sure the circuit will oscillate at this frequency ω at a negative feedback phase shift of ±179° or a positive feedback phase shift of ±359°.
If you meant this, I think you're right!