Out of curiosity I'm trying to design an audio frequency range oscillator that would produce a pure sine wave. I'm working in simscape (simulink - matlab) and here is the design I produced : Schematic

The idea is pretty simple, the left part (without the transistors) is a band-pass filter with a gain of 1/3, which feeds into an op-amp non inverting amplifier. The gain of the amplifier is controlled by a JFET and we feed the output of the amplifier to its gate so the oscillations don't saturate the op-amp. We make use of a voltage divider to feed its gate to have a better control on the amplitude of the output oscillations.

So without the left part everything runs the way I expected it, but I wanted to have a voltage-controlled oscillator, so I decided to use FETs as voltage-controlled resistors put in parallel with the already existing resistors. Then I set a DC voltage between the sources of both transistors and their gates (if later I want to feed a voltage that comes from another circuit I could use a comparator) to control the resistance of the FETs and ensure that both of them have a close resistance.

If I am actually able to produce a voltage-controlled oscillation with this design, adding the FETs added a lot of distortion which I attribute to their non-linearity in the triode region (I triple checked the current flowing through all of them to ensure that they do not saturate), which I do want to reject.

So is there a way on this design to reduce the distortion on my waves ? I am not very used to working with FETs... I was thinking that I could also filter the added harmonics afterwards but then I'd need a Voltage Controlled Filter, and with a similar design I guess that I would see the same distortion so I'm not sure about that solution.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.