RC Phase shift oscillator

[Amirsd]

Hi all
I am new to this forum
I have a question
how can the rc oscillator produces sin wave which component cause it ?
I mean for example in lc oscillator we have lc parallel that they produce sin wave but here there is not something like that to produce sin wave .
I hope I could write my question clearly 8-O8-O8-O

 

[crutschow]

A Phase Shift Oscillator can produce a sine-wave because there's only a single sine-wave frequency that has the proper phase shift through the RC phase-shift circuit feed-back loop to cause an oscillation. Of course to avoid distortion of the signal you need to carefully control the gain of the feedback so the amplitude will not increase to the point of signal clipping. This is often done with some sort of AGC gain feedback circuit (such as by using a FET or incandescent bulb). For Phase-shift oscillators with multiple stages (such as a Bubba oscillator) you may also be able to soft-clamp one of the stages with back-to-back diodes since the subsequent RC stages will filter the squarish edges of the clamped stage to give a relatively low distortion output.

 

[Amirsd]

Thanks for response
So if I do not use feedback line I mean just tree caps and tree res that each make 60 phase shift
and a initial condition for first cap on the output I have a sin wave ? or it is only for feedback effect that sin wave is generated ?

 

[Audioguru]

The three RC phase shift parts create a total of 180 degrees of phase shift which is used for positive feedback and the inverting transistor causes the remaining needed 180 degrees of phase shift. The phase shift parts cause an attenuation of about 29 times.
The voltage gain of a single transistor phase shift oscillator must be limited to a little more than 29 to avoid clipping distortion. Negative feedback is used to limit the voltage gain.

 

[LvW]

I mean for example in lc oscillator we have lc parallel that they produce sin wave but here there is not something like that to produce sin wave .
I hope I could write my question clearly 8-O8-O8-O

It is not true that we have "lc parallel that they produce sin wave".
In any case, we need an active element (amplifier) with feedback which allows to establish a loop gain of unity (Barkhausen´s oscillation condition) at one single frequency only.

In case of an LC oscillator this frequency selection is realized with a bandpass and for other types (RC phase shift) a lowpass or a highpass is used.

 

[nima_1981]

may be this useful for you .


https://s1.picofile.com/file/7280027525/osnew.jpg

 

[BradtheRad]

Thanks for response
So if I do not use feedback line I mean just tree caps and tree res that each make 60 phase shift
and a initial condition for first cap on the output I have a sin wave ? or it is only for feedback effect that sin wave is generated ?

The phase shifts do not need to be equal. The capacitors and resistors can have assorted values. They will find a frequency between them which adds up to 180 degrees total phase shift.

You can also use 4 capacitors.

 

[crutschow]

.......................... or it is only for feedback effect that sin wave is generated ?

If you want an oscillator then you need feedback.

 

[Amirsd]

Hi all thanks for your responds

If you want an oscillator then you need feedback.

My main question is this
I want to give you an example my question is actually due to physical respect of rc respect
for example in lc circuit if you have one capacitor and one resistor parallel to it you do not have any cycles i mean in positive side of cap never negative sign will be created .
but if I replace res by inductor for reason of resonance effect . here magnetic field converts to static and vice versa
so continuing the resonance and creating sin wave . (if there was not resistance feedback is also not required in ideal mode ) so we can get result . this device also with a low resistance can produce at least one cycle of sin wave by lowering amplitude .but when you have tree caps and 4 res like BradtheRad image (without other devices on image just caps and res
and if first cap of the left has initial value you have wave that has positive and negative sign but that is not sin wave it is just a wave with one cycle and we can continue it with feedback I also tested this first step without feedback and get my result
please tell me what is the physical reason that there must be a sin wave instead of other wave mode ?
and if in any piece of my text is any wrong please say .
thanks for helping

 

[Audioguru]

in lc circuit if you have one capacitor and one resistor parallel to it you do not have any cycles i mean in positive side of cap never negative sign will be created.

A parallel LC circuit resonates. Its resonance causes it to be a high impedance at the resonant frequency. If it is fed AC then its capacitor voltage swings positive then negative then positive then negative etc. But you can also feed it AC with DC on it then its DC voltage swings more then less then more then less etc.

but if I replace res by inductor for reason of resonance effect . here magnetic field converts to static and vice versa

You make no sense.

if there was not resistance feedback is also not required in ideal mode

Please do not talk about feedback unless you say if it is positive feedback to make an oscillator or if it is negative feedback to reduce the gain of an amplifier.

this device also with a low resistance can produce at least one cycle of sin wave by lowering amplitude

You make no sense.

but when you have tree caps and 4 res like BradtheRad image (without other devices on image just caps and res
and if first cap of the left has initial value you have wave that has positive and negative sign but that is not sin wave it is just a wave with one cycle and we can continue it with feedback I also tested this first step without feedback and get my result

The four RC phase shifts add up to 180 degrees of phase shift (three RC phase shifts can also add up to 180 degrees) and the transistor is inverting so it also produces 180 degrees of phase shift.
When the output is fed to the input then the total feedback is 360 degrees of positive feedback. Then the circuit oscillates at the frequency that the RC phase shifts add up to 180 degrees.

Negative feedback can be added to reduce the voltage gain of the transistor so that its output has no clipping.

please tell me what is the physical reason that there must be a sin wave instead of other wave mode ?

When it is not clipping then the output is a single frequency. A sinewave is a single frequency. A squarewave, triangle wave and sawtooth wave have many additional harmonic frequencies.

if in any piece of my text is any wrong please say .

I speak, read and write only in English. Your English and spelling are pretty bad but I think I understood about half of your text.

 

[Amirsd]

I speak, read and write only in English. Your English and spelling are pretty bad but I think I understood about half of your text.

ُSorry if I can not express my question better
I don't think that my English is too bad that only half of my text is Intelligible
you have wrote on some pieces of my text :

You make no sense.

if your mentions that it is not Intelligible (I do not know the mean of this sentence my native language is not English )
I think this sections is very clear :

but if I replace res by inductor for reason of resonance effect . here magnetic field converts to static and vice versa

if there was not resistance feedback is also not required in ideal mode

this device also with a low resistance can produce at least one cycle of sin wave by lowering amplitude

The four RC phase shifts add up to 180 degrees of phase shift (three RC phase shifts can also add up to 180 degrees) and the transistor is inverting so it also produces 180 degrees of phase shift.
When the output is fed to the input then the total feedback is 360 degrees of positive feedback. Then the circuit oscillates at the frequency that the RC phase shifts add up to 180 degrees.

When it is not clipping then the output is a single frequency. A sinewave is a single frequency. A squarewave, triangle wave and sawtooth wave have many additional harmonic frequencies.

You are always taking about phase and positive and negative feedb I know them .
First you say wave is Sinusoidal then you proof it again .
Here I can not see any physical reason, for this wave to be Sinusoidal .
Please just say me the physical reason of this phenomenon . like :
https://en.wikipedia.org/wiki/LC_circuit

 

[crutschow]

Okay, let's try again. A simple resonant circuit resonates at one frequency. For example, an LC circuit is resonant at the one frequency where the inductive reactance equals the capacitance reactance. This one frequency is a sinusoid since any other waveshape would have harmonics of the frequency (from Fourier analysis) and these frequencies are not at the resonant frequency. Frequencies not at the resonant frequency are suppressed due to the steep slope of the impedance change of a resonant circuit with frequency (for example, it's zero ohms at the resonant frequency of an ideal LC series circuit or infinite ohms for a parallel LC circuit, and rapidly changes as the frequency goes away from resonance).

If you have a problem seeing all this than I suggest you simulate some of these circuits on a Spice simulator such as the free LTspice simulator from Linear Technology. With that you can see the various voltages and currents in a circuit, to observe the affects of resonance and the phase shifts that occur in an oscillator.

If you are looking for a physical reason for all this it's the change of impedance with frequency of a capacitor and inductor. If you don't understand that then you may need to study more about inductors and capacitors.

 

[Audioguru]

You cannot replace the phase shift resistors in a phase shift oscillator with inductors because then you will have a series LC resonant circuit to ground which has an extremely low impedance at its resonant frequency. Then the signal is shorted to ground.

 

[The Electrician]

ُHere I can not see any physical reason, for this wave to be Sinusoidal. Please just say me the physical reason of this phenomenon.

The physical reason is that the operation of the circuit is described by an ordinary differential equation with constant coefficients.

The solution of such differential equations is always a combination of exponentials. If the exponentials have imaginary parts in their exponents, that gives a sum of cosines and sines, possibly multiplied by decaying real exponentials.

See:

http://sosmath.com/diffeq/second/constantcof/constantcof.html

 

[LvW]

The physical reason is that the operation of the circuit is described by an ordinary differential equation with constant coefficients.

Yes - and going further "backwards": Remember the fundamental laws: V=L*dI/dt and I=C*dV/dt