Re: Random thoughts on sinewave oscillators

On 10/20/24 2:55 AM, piglet wrote:

KevinJ93 <kevin_es@whitedigs.com> wrote:

On 10/19/24 12:57 PM, piglet wrote:

KevinJ93 <kevin_es@whitedigs.com> wrote:

On 10/19/24 12:13 PM, piglet wrote:

On 18/10/2024 4:46 pm, john larkin wrote:

On Fri, 18 Oct 2024 11:25:19 -0400, "Edward Rawde"
<invalid@invalid.invalid> wrote:
>

"piglet" <erichpwagner@hotmail.com> wrote in message
news:vetde5$38sbk$1@dont-email.me...

Edward Rawde <invalid@invalid.invalid> wrote:

>
The circuit below produces a reasonable looking sinewave but the rise
time still seems to be slower than the fall time. It may be that the
amplifier in use is not ideal for this.
>

>
Could that just be second harmonic distortion? You could test the
amplifier
by uncoupling the Wien network and injecting test inputs.
>
Elsewhere I think your amplitude control problems could be simply
due to
too much gain.

>
Perhaps, but I've not so far been able to get the circuit I posted in
response to Bill to produce a sine wave no matter what I do
with the control loop gain.
It either grows to clipping or dies.
>

>
>
-- piglet

>

>
I do a lot of instant-start LC oscillators as the timebase of
triggered delay generators. I let them clip just a bit to stabilize
amplitude.
>

>
>
Here is an amusing oscillator that has a voltage follower as the active
stage - it has no voltage gain so some people say it cannot work - they
are wrong of course.
>

<...>

piglet
>

>
If you move the ground to the emitter of Q3 and slide R8 through the
power supply to the collector of Q3 you can see that it is a
conventional phase shift oscillator with feedback from the output of an
inverting amplifier via a 3-stage RC network.
>
kw
>
>

>
What do you do with the cold end of C1 in that scenario?
>
>

>
That connects to the collector, as it is now and forms the first section
of the RC network. (voltage sources are zero impedance from the point of
view of AC). C4 is a coupling capacitor - its value and the high
impedance of R4,R5 and Q1 mean it has little effect on the phase shift.
>

 It still works if I replace the transistors with an op amp non inverting
follower - no inversion in sight. How would you re-describe that?
 

Exactly the same argument applies - any current coming out of the opamp must come from the power supplies and so goes through R8. The gain to the PS pin can be much more than unity.
Some circuit designs exploit this feature - I think I have seen John Larkin describing one. There are also designs in The Art of Electronics using it.