On this page:

http://www.allaboutcircuits.com/vol_3/chpt_9/4.html

a phase shift oscillator is shown with some text that says the phase shift of each section of the CR network is 60°. This is not true, as is mentioned here:

http://wps.prenhall.com/chet_paynter_introduct_6/0,5779,426330-,00.html

Consider the three cascaded CR sections without the transistor circuitry. There is some frequency where the phase shift through the full unloaded network, from the left end of C1 to the right end of C3, is 180°; denote that frequency by ωo.

Sharpen your analysis skills by calculating the attenuation and phase shift in each section at ωo.

What is the actual phase shift through each section; how far is it from 60°?

 

The phase-shift oscillator is too simple so it clips its output signal. Then it is not a sine-wave.

 

That may be, but this thread isn't about the oscillator per se. It's about the phase shift in each section of the 3 section CR network.

 

My phase-shift oscillator is made with a quad opamp. Each RC section is buffered. The first opamp clips the signal a little to establish the output level then the low distortion output is from the 3rd opamp that has been filtered 3 times.

 

Take a look here
http://forum.allaboutcircuits.com/showthread.php?t=26673. I have posted some useful information for you.