Greetings everyone!

I have been poring over my textbooks as well as reading online (this great website included), but I am still very confused when I try to determine the output phase/polarity that an op-amp is going to produce specially in response to positive feedback oscillators, such as the Wien-bridge oscillator.

Suppose, I take an op-amp, give it a direct positive feedback on its (+) input, and a positive voltage signal on its (-) input. Here's what I think will happen: The op-amp will generate an output signal, amplified and reversed in polarity to the input signal, go back around to the (+) input, where it will be subtracted from the signal at the (-) input. But this subtraction will actually result in a reinforcement of the signals, since they are opposite in polarity and op-amps respond to the differences of their inputs, not the actual inputs, right?

Another thing I want to ask: What does the 0 phase-shift requirement really mean in relation to op-amp oscillator circuits? The op-amp takes the difference of the input, so in order to have the input reinforced, shouldn't there actually be a 180 phase-shift around the loop, so that when the op-amp takes the difference, the signal will be reinforced?

If I am correct so far, then could the simple circuit that I have described above perform as an oscillator, or at least meet the 0 phase-shift requirement? I think the latter, at least, is true, and if I am right, why do we need oscillator circuits like Wien-bridge to ensure the 0 phase-shift condition? When I try to follow the polarities around the two feedback loops of a Wien-bridge, I always end up hopelessly lost.

 

To me 0 phase-shift requirement means that input and output signals have same phase. Meaning that when you put input signal on Channel 1 of oscilloscope and output signal on Channel 2 of oscilloscope, you will see that both signal have peaks and troughs in identical places, which means that they are in phase, phase-shift is 0.

 

To me 0 phase-shift requirement means that input and output signals have same phase. Meaning that when you put input signal on Channel 1 of oscilloscope and output signal on Channel 2 of oscilloscope, you will see that both signal have peaks and troughs in identical places, which means that they are in phase, phase-shift is 0.

Yes, I agree. That is the way it is normally understood. But with op-amps configured as oscillators, I am always left puzzled what constitutes an input, because oscillators usually don't have any external inputs coming into them, so which is the input signal? The feedback going into (+), or the one going into (-), or their difference?

 

Yes, I agree. That is the way it is normally understood. But with op-amps configured as oscillators, I am always left puzzled what constitutes an input, because oscillators usually don't have any external inputs coming into them, so which is the input signal? The feedback going into (+), or the one going into (-), or their difference?

Depends on oscillator. Some have external input, some do not. Wien-Bridge has external input.

 

To me 0 phase-shift requirement means that input and output signals have same phase.

Depends on the type of oscillator, and which input. For a Wein bridge oscillator, the non-inverting input and the output never have the same phase. For a phase shift or bubba oscillator, the non-inverting input is tied to a reference potential and has no signal on it. The inverting input is, well, inverted with respect to the output.

ak

 

If you put positive feedback with 0 phase shift at DC, you will have a latch.

The 0 phase shift applies to AC and the phase shift depends on frequency. For a Wien bridge, the feedback will be to the inverting input, and the phase shift will be 180 degrees at the oscillation frequency. Since the inverting input has 180 phase shift from the output, the actual feedback is 180 + 180 = 360 = 0 or 18 - 180 = 0.

Bob

 

Nope. For a Wein bridge oscillator, there is feedback to both the inverting and non-inverting inputs. It is the positive feedback that has the phase shift networks in it. The feedback to the inverting input usually is purely resistive, has no frequency dependent components and no phase shift, and is what sets the circuit gain to sustain oscillation.

https://en.wikipedia.org/wiki/Wien_bridge_oscillator

ak

 

Sorry, you are right. I was thinking of a phase shift oscillator.

Bob

 

Always liked the phase-shift oscillator. Reasonable performance with very little effort, good amplitude stability, tunable over a small freq range with only one resistor. In the 70's I did the reference tone oscillator for a TV station with one. One trimpot to tune it to exactly 400 Hz, one trimpot to set the output to +8 dBm, sooo much easier than a Wein bridge. As of 2004, still in use.

ak

 

Thanks everyone for your help. Appreciate it.