See

@Bordodynov

Hola Alex,

Always wrt your circuit:



After several hours struggling to get it working, found that to get it oscillating I needed to reduce R3 to about 220 Ohms.. Frequency is in the ball park on account I soldered approximate values for C2 & C3. I also made C1 variable and works OK





My questions:

How do you decide if R3 is necessary and how do you calculate its value?

/Edit Please disregard this question; it is not telling what I want to ask. I will pose it again later: What criteria to follow for calculating the impedance of the series C2 + C3? /Edit

Once the above is known, what is the criteria to calculate the divider vis a vis R4 and C1?

Considering to start a similar design afresh, where should I start?

Do you know any place where I could find the whole sequence of steps to calculate this from scratch? Everybody seems in love with analysing but nobody seems akin to explain the calculation sequence in detail.

Sorry Alex for the so many questions but wildly tweaking an existing design does not seem the way I would learn much. Thanks for your time and help.

 

The equivalent capacitance of these capacitors determines the operating frequency, and their ratio determines the value of positive feedback. In order for the output signal distortion to be small, the feedback value must be minimal. This can be conveniently controlled with an additional resistor. The actual circuit has parasitic inductances and capacitance. I used a resistor because it's convenient to adjust the generation conditions. Of course you can do that with capacitors, but it's not convenient - the frequency is also shifted. You did the right thing by selecting this resistor. I didn't count on this resistor, I just picked it up. All the more reason to calculate without taking into account parasitic inductances and capacitors is too optimistic. Plus, there's a variation in the transistor parameters.

 

The equivalent capacitance of these capacitors determines the operating frequency, and their ratio determines the value of positive feedback. In order for the output signal distortion to be small, the feedback value must be minimal. This can be conveniently controlled with an additional resistor. The actual circuit has parasitic inductances and capacitance. I used a resistor because it's convenient to adjust the generation conditions. Of course you can do that with capacitors, but it's not convenient - the frequency is also shifted. You did the right thing by selecting this resistor. I didn't count on this resistor, I just picked it up. All the more reason to calculate without taking into account parasitic inductances and capacitors is too optimistic. Plus, there's a variation in the transistor parameters.

Quite crear Alex
Now, what about the sequence when starting from scratch?