RC phase shift oscillators

Thread Starter

Liz91

Joined May 24, 2012
1
Hi

Im stuck on the question below.

design a phase shift oscillator which gives two outputs at 120 degrees to each.

Im confused as to how many rc stages are required and why?
 

jegues

Joined Sep 13, 2010
733
Hi

Im stuck on the question below.

design a phase shift oscillator which gives two outputs at 120 degrees to each.

Im confused as to how many RC stages are required and why?
The number of stages will be determined by the amount of phase shift at the output of each stage.

For example, if one RC stage provides you with a phase shift of say 60°, then you would want to implement two stages to achieve the desired 120°.

The amount of actual phase shift depends upon the chosen values of R an C, as well as the desired frequency of oscillation.

\(\phi = arctan(\frac{X_{c}}{R})\)
 

Wendy

Joined Mar 24, 2008
23,415
There is a built in assumption though. The amplifier stage inverts the signal 180°. With BJT amplifiers this is a common configuration, as well as with vacuum tubes and other types of transistor amplifiers.
 

Ron H

Joined Apr 14, 2005
7,063
There is a built in assumption though. The amplifier stage inverts the signal 180°. With BJT amplifiers this is a common configuration, as well as with vacuum tubes and other types of transistor amplifiers.
I gotta point out that not all amplifiers invert the signal.:D
 

Wendy

Joined Mar 24, 2008
23,415
In which case a 3X 60° phase shift networks will not work. The 180° is a usual requirement for most phase shift oscillators, as a more specific rule is all the phase shifts must add to a full 360° for positive feedback. Using a amplifier that has 180° simplifies the math enormously, and the 3X 60° network is also a frequency selective component.

Using an amplifier that does not invert the signal for an amplifier is possible, but the design is completely different, and it tends to be a totally different type of oscillator, such as a astable multivibrator that uses a totally different frequency selective principle.

The Bubba oscillator uses a 90° phase shift principle that performs much the same way.

TI has an excellent PDF abstract on the subject.

Sine Wave Oscillators
 

Wendy

Joined Mar 24, 2008
23,415
I've never heard of it. There are lots of things I've never studied though.

A Wien Bridge would come close to describing it too, I suspect. This particular circuit is what gave HP their start, way back when.

I just reread the OPs requirement, we have borderline hijacked her thread.

I would use either the 60° or 120° oscillator that Ron or I described with a very high impedance amp to tap into the appropriate part of the phase shift feedback circuit.

When you add all the various phases they add up to 360°. This provides the required positive feedback for the oscillator to work.
 
Last edited:

atferrari

Joined Jan 6, 2004
4,764
What worked for me at the first try, was a "twin T" oscillator using a TL072 opamp.

Used two 1N4148 diodes to control amplitude and seemed to have very little distortion even if assembled in a protoboard.
 

Wendy

Joined Mar 24, 2008
23,415
What worked for me at the first try, was a "twin T" oscillator using a TL072 opamp.

Used two 1N4148 diodes to control amplitude and seemed to have very little distortion even if assembled in a protoboard.
I suspect this is a Wien Bridge, in the absence of a schematic.

Do not forget the OP wants a 120° phase shifted sine wave. It was part of the requirements. I missed it 1st go around.

There are many ways to do this, the classic 3RC phase shift schematic is just the most common.
 

daviddeakin

Joined Aug 6, 2009
207
Hi

Im stuck on the question below.

design a phase shift oscillator which gives two outputs at 120 degrees to each.

Im confused as to how many rc stages are required and why?
You can use an ordinary 3-stage oscillator. The signal across each stage is 60 degrees away from the ones either side, so you can tap off the signals from the 1st and 3rd stages (with high impedance buffers) to get two signal 120 degrees apart, albeit with different amplitudes.
 
Top