Phase Shift Oscillator Design Help

[godevilrays134]

Hello, I am trying to design a phase shift oscillator using three RC filters. I have chosen the capacitors to be 0.001 μF, the resistors to be 10 KΩ and so the feedback resistor has to be 290 KΩ. With these parameters, the oscillation frequency should occur at about 6.5 KHz. When I simulate this circuit however, I do not observe a 180 degree phase shift at this frequency rather I see that the original and the output wave appear to be perfectly in phase. I am wondering if I am placing my function generator in the wrong location and if I should be using a resistor with the function generator. Attached is a schematic of the design as well as the input and output signals on an oscilloscope. Any suggestions would be appreciated. Thank You

 

[john monks]

Where is your original signal and the output signal?

Normally we don't talk about an input to an oscillator.

If you are talking about the signal from your function generator and your oscilloscope then your circuit is doing as is expected. The operational amplifier inverts the signal 180 degrees and R1 C1 R2 C2 R s C3 shift the waveform another 180 degrees. So the function generator and the oscilloscope are seeing a total of 360 degree phase shift as is expected.

One thing I'm confused by. How can you have such a large signal on the inverting input of an operational amplifier? This input should be about zero ohm impedance. Maybe you should place a resistor in series with your function generator and they again.

 

[godevilrays134]

Where is your original signal and the output signal?



If you are talking about the signal from your function generator and your oscilloscope then your circuit is doing as is expected.

First and foremost, thank you for the prompt response. Those were indeed the signals I was referring to. I'm glad to know that the 360 degree phase shift is actually what is suppose to occur. I will add a 1 KΩ resistor in series with the function generator. So if I'm understanding the circuit correctly, it should be the case that at frequencies below the inversion frequency (6.5 KHz) the phase shift between the original signal and the output signal will be close to 0 but at frequencies above 6.5 KHz the phase shift will approach -90 degrees asymptotically. Why is it exactly that we generally don't show the AC source when we discuss oscillators?

 

[john monks]

An oscillator generates it's own signal. It doesn't normally require a source. Now because you are using a computer simulation you will need ti tickle the oscillator into oscillation, something that is not necessary in a real circuit.

Assuming the inversion is about 6.5 kHz, it looks right, then at higher frequencies your phase shift becomes less than 360 degrees. Below 6.5 kHz the phase shift is more than 360 degrees.

I notices that the battery going to pin 4 is reversed.

 

[godevilrays134]

An oscillator generates it's own signal. It doesn't normally require a source. Now because you are using a computer simulation you will need ti tickle the oscillator into oscillation, something that is not necessary in a real circuit.

Assuming the inversion is about 6.5 kHz, it looks right, then at higher frequencies your phase shift becomes less than 360 degrees. Below 6.5 kHz the phase shift is more than 360 degrees.

I notices that the battery going to pin 4 is reversed.

I have corrected the battery going to pin 4 so that it is now -15V. Using AC analysis I have made a bode plot of the phase shift between the input and output signals (see attached). I do not understand why the phase shift jumps from -180 degrees to 180 degrees at 40 KHz.

 

[duke37]

Phase jump.
It is just to fit on the graph. Once round the circle is the same as starting again.