1
\$\begingroup\$

I am trying to build a "quasi" square wave oscillator means the output voltage should be positive only and edges at least little sharp. I know it it possible to use a comparator connected to sine and make it nicely but I would like to avoid it.

One think I can imagine is to place resistor directly to collector circuit what forms the output to square and only little part of output voltage carry to base LC tank (using high gain transformer).

So in numbers if Vcc=3.5v the most voltage should be on R2 (about 3v) and some small part of it on L1 (0.5v).

enter image description here

I tried to build that circuit and I got something very close to my expectation:

Voltage at collector: enter image description here

Voltage across LC tank: enter image description here

But what I do not like is shape of Ic current, I think it is not symetrical enough what makes not symetrical output voltage also. Focus on little decrese at the top of waveform in picture below. It is the place when the Ic cannot rise further becouse of not enough Ib.

Ic - collector current (or another words voltage across R2) enter image description here

enter image description here

Does anyone know how to handle this top to be symetrical, what happened there? I tried to simulate it and I cannot find a soultion, maybe it is not even possible. The L1 inductor has maximum energy at that place, so only way it can flow is into capacitor in base circuit.

\$\endgroup\$
3
  • \$\begingroup\$ Analyzing this circuit isn't something I'm into doing ATM. But I'll make an observation. Your scope probe is going to be a significant electrical thing to this circuit - i.e. the act of measuring may very well be influencing it's operation. Consider the input impedance (resistance and capacitance) of your scope probe.... When I see circuits with resistors of 1k-100k, I don't consider it. But 400k + 33nF??? Those values are not enormously far from the values of some scope probes. What are yours spec'ed at? \$\endgroup\$
    – Kyle B
    Sep 1, 2021 at 5:54
  • \$\begingroup\$ @Kyle B Yeah, my probe is 1M ohm. I am realizing the measuring of base waves can be affected comparing to 400k bias resistor. But the collector values must be good. \$\endgroup\$
    – user208862
    Sep 1, 2021 at 5:59
  • \$\begingroup\$ Trying to make a square (ish) wave oscillator using an LC tank is naïve - use a relaxation oscillator - that is what they are intended for. \$\endgroup\$
    – Andy aka
    Sep 1, 2021 at 7:20

1 Answer 1

1
\$\begingroup\$
  1. I would put some impedance (R || C) between Emitter & GND.
  2. Tab out the collector and re-shape the signal.
  3. LC oscillator can be built using logic gates, an example here.

enter image description here

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.