Phase shift Oscillator

[letitbleed]

When I put 0.00001 volts the amplitude decreases 😞 Why

[jg_geadinc · Accepted solution]

You have the gain set with the pot such that there is just barely enough feedback for it to sustain oscillation. If you increased the value of the pot it will grow. It will probably grow as you have it set now, you just have to wait long enough. 

[letitbleed]

Wise, indeed, very wise. This has more tricks to it than just component values. Definitively, working at its best just in 0.001 uV and playing with the pot. to push it. Sorry if I am being to buggy but where I live/study there aren't many experienced people in this area to ask stuff...  Is there a way of doing an AC analysis in this? I once did one in a Butterworth filter but can't seem to be able to do it again

[jg_geadinc]

You can, but you would need to open the loop. You can do things like leave the 1k connected to the inverting input, sweep that with a function generator and examine the output at the end of the phase shift path. You can then see the phase shift as a function of frequency. Where the phase shift is 180 degrees should be the same as the frequency it oscillates at in your current file. If you open the circuit at the output and drive it through the phase shift network through the RC chain you can see how much gain the circuit has and how it changes with the pot and observed if the value of the pot has any influence on the phase shift and therefore the frequency of oscillation. Analyzing an oscillator by opening up the feedback loop is how RF VCOs are analyzed using software such as Genesys. It allows you to analyzer the Q of the resonator, the gain of the oscillator, the phase shift, etc. 

[letitbleed]

Hello again. Why does anything less than 0.001 uV makes the amplitude actually decrease? Is it not enough impulse?

[jg_geadinc · Accepted solution]

I reading this on my phone so I can't try it to see. It's probably not enough to overcome the offset voltage on the input of the opamp.
Try increasing the values of all the resistors by 10x and reducing the capacitors by 10x. The frequency will be the same but the amount of current in the feedback loop will be 10x smaller.

[letitbleed]

How is the gain of this oscillator determined anyhow? Physically I get different values than in the simulation

[letitbleed]

Do you know by any chance why doesn't my simulation in Simulink doesn't work?

[jg_geadinc]

The DC gain is Rf/Rin which in your case is 290k/10k+10k+10k+10k = 7.25. 

[jg_geadinc]

Can't help you with Simulink, never used it before.