Maker Pro
Maker Pro

Oscillation amplitude of a opamp sine wave oscillator

B

Bill yg

Jan 1, 1970
0
Dear All,

Perhaps this is a dumb question but I really find confused.

I am building a simple RC-phase shift oscillator using an opamp, with
+/- 15V power supplies. I used a trimmer in the feedback resistor
across the opamp. However, I found something interesting.

The output amplitude is normally +/- 15V. The waveform is somewhat a
fine-looking sine wave.
If I set a larger opamp gain, the output waveform is distorted and
looked like a square wave.
If I set a smaller opamp gain just enough to start oscillation, it
will result in a +/- 12V or smaller sine wave.

The question is:
Why is this so? Why won't the +/- 12V sine wave grow up to +/- 15V?
What I mean is, in theory, the oscillation should grow up to
infinity... and so why now is bounded by +/- VCC? and even with a
smaller gain, bounded by +/- 12V?

Thank you very much.

Billy yg
 
R

Robert Baer

Jan 1, 1970
0
Bill said:
Dear All,

Perhaps this is a dumb question but I really find confused.

I am building a simple RC-phase shift oscillator using an opamp, with
+/- 15V power supplies. I used a trimmer in the feedback resistor
across the opamp. However, I found something interesting.

The output amplitude is normally +/- 15V. The waveform is somewhat a
fine-looking sine wave.
If I set a larger opamp gain, the output waveform is distorted and
looked like a square wave.
If I set a smaller opamp gain just enough to start oscillation, it
will result in a +/- 12V or smaller sine wave.

The question is:
Why is this so? Why won't the +/- 12V sine wave grow up to +/- 15V?
What I mean is, in theory, the oscillation should grow up to
infinity... and so why now is bounded by +/- VCC? and even with a
smaller gain, bounded by +/- 12V?

Thank you very much.

Billy yg

Op amps at best, cannot go beyond the supply voltages; that should be
obvious.
The non-rail-to-rail opamps normally have problems going within about
3V of the positive supply and within 1V of the negative supply.
If there is too much gain, then the opamp simply clips the peaks
because it cannot go beyond the supplies.
If you do not have enough gain, it *cannot* oscillate!
If you get a +/- 12V sine wave at some gain settings, that is an
indication you have some nonlinear amplitude limiter in the circuit.
 
B

Bill yg

Jan 1, 1970
0
Robert Baer said:
Op amps at best, cannot go beyond the supply voltages; that should be
obvious.
The non-rail-to-rail opamps normally have problems going within about
3V of the positive supply and within 1V of the negative supply.
If there is too much gain, then the opamp simply clips the peaks
because it cannot go beyond the supplies.
If you do not have enough gain, it *cannot* oscillate!
If you get a +/- 12V sine wave at some gain settings, that is an
indication you have some nonlinear amplitude limiter in the circuit.

Thanks for your reply. So, is this circuit-related? I mean, there is
no mathematics that governs why at a particular gain, the output is
+/- 12V? Or even there IS mathematics, the equations themselves vary
from one opamp to the other?
 
R

Robert Baer

Jan 1, 1970
0
Bill said:
Thanks for your reply. So, is this circuit-related? I mean, there is
no mathematics that governs why at a particular gain, the output is
+/- 12V? Or even there IS mathematics, the equations themselves vary
from one opamp to the other?

The math behind the required gain and phase for oscillation has
nothing to do with the op-amps.
The op-amps are the means of realizing gain and phase reversal (if
needed).
So, start with a phase-retard oscillator circuit: "source" to R (TP1),
R from TP1 to TP2, R from TP2 to TP3; C from TP1 to ground, C from TP2
to ground, C from TP3 to ground; gain and phase reversal stage from TP3
to "source". I think the minimum gain is about 39 (too lazy to
calculate); the waveform at TP3 is close to a sine wave (3% distortion)
if "source" is a square wave.
Amplitude is limited by the gain and phase reversal stage limiting as
i mentioned about op-amps.
Your 12V output level comes from some kind of non-linear device in
your circuit: zeners, incandescent lamp, FET as either gain adjust or
amplitude adjust, an op-amp limiting.
Cannot say what, as i know nothing about your circuit and what parts
you are using.
 
B

Bill yg

Jan 1, 1970
0
Robert Baer said:
The math behind the required gain and phase for oscillation has
nothing to do with the op-amps.
The op-amps are the means of realizing gain and phase reversal (if
needed).
So, start with a phase-retard oscillator circuit: "source" to R (TP1),
R from TP1 to TP2, R from TP2 to TP3; C from TP1 to ground, C from TP2
to ground, C from TP3 to ground; gain and phase reversal stage from TP3
to "source". I think the minimum gain is about 39 (too lazy to
calculate); the waveform at TP3 is close to a sine wave (3% distortion)
if "source" is a square wave.
Amplitude is limited by the gain and phase reversal stage limiting as
i mentioned about op-amps.
Your 12V output level comes from some kind of non-linear device in
your circuit: zeners, incandescent lamp, FET as either gain adjust or
amplitude adjust, an op-amp limiting.
Cannot say what, as i know nothing about your circuit and what parts
you are using.

Thank you very much !
 
Top